EM-1 Capabilities of Nuclear Weapons

LA-4756

c 3

CIC-14 REPORT COLLECTKIN

REPRODUCTION
COPY

Proceedings

of
Environmental
Plutonium
Symposium

I
Held
at
LASL,
August
4-5,
1971

Iamos

scientific
laboratory

of the University
of California

LOS ALAMOS,
NEW
MEXICO
87544
[
\

UNITED
STATES

ATOMIC
ENERGY
COMMISSION

CONTRACT
W-740
S-ENG.
36

a

ad

This report was prepared as an account of work sponsored by the United

States Government.
Neither the United States nor the United States Atomic

Energy
Commission,
nor any of their employees, nor any of their contrac-

tors, subcontractors, or their employees, makes any warranty, express or im-

plied, or assumes any legal liability or responsibility for the accuracy, com-

pleteness or usefulness of any information,
apparatus, product or processdis-

closed, or represents that its use would not infringe privately owned rights.

This report expressesthe opinions of the author or authors and does not nec-

essarily reflect the opinions or views of the Los Alamos Scientific Laboratory.

Printed in the United States of America.
Available from

National Technical 1nformation
Service

U. S. Department of Commerce

6285 Port Royal Road

Springfield,
Virginia
22151

Price:
Printed Copy $3.00;
Microfiche
$0.95

*

.

[0

(‘

lames

LA-4756

u C-41

ISSUED:
December 1971

.

scientific
laboratory

of
the
University
of
California

LOS ALAMOS, NEW MEXICO 87544

Proceedings

of
Environmental
Plutonium
Symposium

Held
at
LASL,
August
4-5,
1971

by

Eric
B.
Fowler

Richard
W.
Henderson

Morris
F.
Milligan

PROCEEDINGS
OF

ENVIRONMENTAL
PLUTONIUM
SYMPOSIUM

held at the

Los Alamos Scientific Laboratory

of the

University of California

Los Alamos, New Mexico

August 4 and 5, 1971

FOREWORD

The
purpose
of
this symposium
was to
discuss the
distribution
and

measurement of plutonium
in the environment.
To this end, the subject matter

has been divided into three broad categories, the first dealing with distribution

or
how
plutonium
has entered
the
environment,
the
second dealing with

methodology
or the means by which one obtains environmental
samples and

analyzes them,
and the third
with the results obtained
from
such measure-
ments and the interpretation which can be inferred from them.

Eric B. Fowier

Richard W. Henderson

Morris F, Milligan

Cochairmen

.

WELCOMING REMARKS

by

Harold M. Agnew,
Director,
LASL

I am delighted
to welcome you all here this morning. When we fust talked about the possibility
of having this meeting,
it was thought
that
there might be twenty
or thirty
people who would be
interest ed and would
come out for the kick-off symposium
on this particular
subject. As you can see,
attendance
has escalated in an exponential
fashion. As you are aware, we’ve been involved here at Los
Alamos with plutonium
for a long, long time. In the beginning the plutonium
as a nitrate
came from
Hanford.
We had the task of putting
it in metallic
form and developing
the metallurgy.
As you are
aware, the first weapns
were actually fabricated
here. The basic plutonium
chemistry
and metallurgy
had to be developed
and carried out. We had a very large building called “’D” Building which we have
somehow
enviornmentally,
I hope, disposed
of - I sometimes
wonder
how we ever did what we did
then.
I have
a feeling
it wouldn’t
pass today.
I certainly
know
that,
when
one
thinks
of the
experiments
we used to do, not only in Nevada and the Pacif3c but right here, take the RaLa work in
particular,
I believe we wouldn’t
have a snowball’s
chance
in hell of doing the things we used to
do - and we thought
we were being very prudent,
being very careful - and, of course, since we lived
here, had a personal stake in what we did.
I think we took all possible,
at least in the context
of those days twenty
or twenty-five
years
ago, prudent
precautions.
As those of you who are now in the business are fully aware, we are today
in a completely
new ball park. I think it is probably justified.
Sometimes,
however, we have a feeling
that
people
are going a little
bit overboard
in the publicity,
and types
of hysteria
that goes with
certain
types
of publicity,
perhaps
more
to get attention
than
to express
legitimate
concern
in a
technical
or medical sense. But nevertheless,
we are very concerned,
as I mentioned,
not only because
of the overall impact
on the environment,
but because
we actually
live here. You will probably
find
more Sierra Club members
in fact or in spirit, per you name it, in Los Alamos than any other city or
institution
in the United
States.
So we are persomlly
very much involved. Our friends from Rocky
Flats,
whom
I see here, many
of whom
came from
here, and were here in the original days -- Bill
Bright, and Ed Walko, and many of the other
people who left here - know what I mean. They went
to the Flats and we all know the problems that they’ve had with their plutonium
in the environment.
I think
it behooves
us all to do the very best we can in an objective
manner.
The problems
that are
facing us today
are probably
nothing
compared
with the problems we are going to be faced with ten,
twenty,
thirty,
ftity
years
from now. There
is no question
that
nuclear
power, not only ordinary
fission reactors
but the liquid metal or other
typ
of fast breeders,
are going to be a reality. We are
going to
have all sorts
of problems
with
regard
to
the
disposal
of radioactive
wastes,
low-level
plutonium,
and
fission
products.
Someday,
hopefully,
the
fusion
projects
will come into
being.
Maybe optimistically
it will be thirty
years from now that we will really have an on-line prototype
fusion electrical
power unit.
In the meantime,
and even long beyond
that, we are going to be faced
with
problems
of materials
such as plutonium.
I believe that the work and interest
you people are
involved in at a symposium
such as this are going to lend to, let’s say, an objective,
rational
approach
that
the leaders
of the country
can follow.
In this manner
I believe that the people in the country
who are concerned
will recognize
the use of plutonium
as being in their best interest
and not being
carried
out just for the pleasure of some ‘White coated”
scientists
who really don’t understand
the
problem.
Again, I am delighted
to have you all here and am looking
forward to seeing you this evening.
My best wishes for a very successful meeting. Thank you.

.

.

2

PLUTONIUM
DISTRIBUTION
AS A PROBLEM
IN ENVIRONMENTAL
SCIENCE

.

Introduction

by

W. H. Langham

Biomedical Research Group

University of California

Los Alamos Scientific Laboratory

Los Alamos, New Mexico

ABSTRACT

The
potential
uses of
plutonium
in future
peace-time technology
are

numerous
and if realized will result in a production
rate of thousands of kg.

per year
by
the end of the century.
By the year 2000
it is predicted
that

plutonium
may be producing
50% of the country’s total energy needs, 3 times

the amount of electrical energy now produced
from
coal, gas, oil, hydro, and
nuclear
energy
altogether.
Power
sources for
mechanical
haarts and
heart

pacers alone
will
require
large quantities
of
‘%,
as will
thermoelectric

generators for deep-space missions, space platforms, and communications satel-

lites. The
technology
of
plutonium
production
and
processing is already

established. Whether
plutonium
attains its predicted role in the future
power

economy
may depend entirely on whether economically competitive methods

of preventing its distribution
in the environment
can be attained. Repetition of

the mercury situation cannot be tolerated although,
in some ways, plutonium

(by
its chemical nature)
is not as devious as mercury
as a potential general
environmental
contaminant.
Because of its volubility and other characteristics,

it is not readily taken into the ecological chain. No natural bacterial or other
environmental
entity has been observed that converts plutonium
to a solubil-

ized form
that readily enters the ecological cycle; however, this possibility is
worthy
of
further
investigation.
Control
of
plutonium
as an environmental

contaminant
involves control
of distribution
from
production
reactors, proc-
essing plants, storage sites, and inadvertent releasesduring transportation
and

use. An
ail important
factor in the alleviation of plutonium
distribution
as a
problem
in environmental
science is continuous surveillance with sensitive and

standardized
methods
of
monitoring
not only
operational
discharges but en-
vironmental distribution
as well, which is the theme of this conference.

........... .. .............. ..... .. ......... ... .............

In his welcoming
remarks,
Dr. Harold
Agnew, the
LASL Director,
mentioned
the fact that the problems
we
face
in
dealing
with
radioactive
contamination
of our
environment
are considerably
smaller today than they will
be in the next
two
or three
decades.
There is no better
way of empahsizing
his remarks
than to refer to Table 1,
developed
from a talk entitled
“The Plutonium
Economy
of the Future,”
given by Dr. G. T. Seaborg at the Fourth

International
Conference
on Plutonium
and Other
Acti-
nides on October
5, 1970, in Santa
Fe, New Mexico. Dr.
Seaborg’s
projections
were based in part
on the Federal
Power
Commission’s
predictions
of the
nation’s
future
power requirements
and the increasing percentages
of that
power that will come from nuclear
sources. He visualized
that
the
annual
production
rate
of ‘%
will increase
from
about
20,000
kg
in
the
1970-1980
period
to
60,000 kg in the 1980-1990
decade,
and to 80,000
kg in
the
period
1990-2000.
Based
on
current
trends
in the

3

PLUTONIUM

Plutonium-239
Power
Production

Plutonium-238
Space
Applications
Medical
Applications

Transplutonium
Isotopes
Curium-244
Crdifomium-252

TABLE
I

ECONOMY
OF THE
FUTURB*

Annual
Production
and/or
in Use
(kg)

1970-1980
1980-1990
1990-2000

20,000
60,000
80,000

10-20
100
..
.,
5
6,000

40
180
200
0.1
0.8
3.5

*G. T. Seaborg (October S, 1970).

space program and visualized applications
in the biological
and medical
field, he postulates
the rate
of production
and use of au
could increase from
10 to 20 kg in the
1970-1980
period
to
100 kg
in
1980-1990
with
the
amount
in
use in power
sources
for mechanical
heart
pumps
reaching
perhaps
6000 kg near
the
turn
of the
century.
This is a staggering
amount
of ‘Pu
when one
‘%% equivalents
by multiplying
by a
puts it in terms of
factor
of -270,
the ratio of their specific activities.
Dr.
Seaborg visualizes also that the production
rate and utili-
zation
of the transplutonium
isotopes
of ‘Cm
and ‘s~f
could
reach
200 and 3.5 kg/yr,
respectively,
by the year
2000.
These
are not inconsequential
amounts
of radio-
activity
when one considers
that the half-life of %m
is
18 yr and ‘~f
is 3.5 yr. As the subject of this conference
is directed
toward
methods
of quantitating
plutonium
in
the environment,
no further
consideration
need be given
to these
latter
materials.
To appreciate
more
fully Dr.
Seaborg’s plutonium
economy
of the future, a little more
discussion might be in order.

Plutonium-239
and Power Production

The
trend
in annual
rate
of production
of ‘%%
reflects,
of course,
the increasing
national
power
needs
over the next
three
decades
before
commercial
thermo-
nuclear
energy
production
may become
a technical
and
economic
reality.
Figure
1 shows
the
Yankee
atomic
electric
station
near Rowe,
Massachusetts,
the first elec-
tric generating
plant built under the AEC’S Power Demon-
stration
Reactor
Program.
Reactors
of this type, the fust
to
supply
commercial
power,
utilize
only
the
‘SU

constituting
approximately
0.7%
of
natural
uranium.
Their
inefficient
utilization
of the
nation’s
natural
re-
sources
of uranium
eliminates
them
as a candidate
for
meeting
the nation’s
expanding
power needs. The current
generation
of
power
reactors
[Light
Water
Reactor
(LWR)]
is based
on
a plutonium+mriched
fuel cycle.
Plutonium
produced
during
operation
is separated
and
added
back to the fuel, resulting in about one-third
of the
total
heat
output
coming
from
plutonium
fission with
production
of more plutonium
for recycling.
This recycl-
ing of the by-product
plutonium
increases the efficiency
of utilization
of the nation’s
uranium
resources
but still
requires
substantial
amounts
of new natural
uranium.
The
next
generation
of reactors
is already
a subject
of exten-
sive research
and
development
by both
the
AEC
and
industry.
This generation
is the Liquid Metrd Fast Breeder
Reactor
(LMFBR)
and
utilizes
the
energy
inherent
in
‘U.
Such a reactor
will breed ‘%
from ‘SU and will
derive about
80% of its energy output
from ‘vu
fission
and the other
2~0
from fast fission of ‘U,
while produc-
ing enough
additional
plutonium
to provide
fuel for new
reactors.
This
progression
of power
output
through
in-
creased
production
and utilization
of ‘~
accounts
for
the increasing rate of production
of the latter as projected
by Dr. Seaborg and concurrently
for its increasing poten-
tial as an environmental
contamination
problem.

Plutonium-238

The
potential
for
production
of
‘%
increases
directly
with increasing
production
of nuclear
power.
In
many
respects
-u
is
an
ideal
fuel
for
reliable

4

I

Fig. 1

thermoelectric
generators
having
~ high- ratio
of power
output
to weight and volume. Such generators
are finding,
and
will continue
to
find,
numerous
novel and unique
applications
as production
capability
and cost
of ‘%%
become more and more favorable.

Space Applications.
Figure 2 shows the fuel capsule
and
graphite
fuel
cask
of the
SNAP-27
thermoelectric
generator.
The fuel capsule
contains
thousands
of curies
of
‘%
in
oxide
form
and
has
an
output
of about
1S00 W of
thermal
power.
Three
of these
devices
are
already
powering
experimental
stations
on
the
moon
(Apollos
12, 14, and 15), and a fourth
(Apollo
13) resides
intact
Other
power

in the
deep
trench
of the
South
Pacific
Ocean.
similar
“%
oxide
heat
sources
are
providing
for
orbiting
weather
and
navigational
satellites.

Undoubtedly
these
applications
will
increase
and
new
ones
will
develop
over
the
next
two
decades
such
as
power
supplies
for
condensers
of biological
wastes
on
long-duration
manned
space missions
and orbiting
space
stations.
Other
foreseeable
space
needs
during
the next
decade
or so are for power supplies on non-manned
plane-
tary
fly-bys
and landings
such as the Grand
Tour
of the
planets
and the Viking
program
already
in the planning
stage.

Biological
and
Medical
Applications.
Some
of the
most novel and intriguing
applications
of 2%% sources are
in the realm of biology
and medicine.
One already begin-
ning
to
be applied
is as a battery
for circulatory-assist
devices,
an example
of which is the heart
pacer (Fig. 3).
In
this
application
each
device
requires
about
0.5 g of

5

FORWARD
CAmU~E
SUPPORT

FUEL

BACK

END
CAP
PRIMARY
HEAT
SHIELD
:,,
..
i:.
.

Rg. 2

plutonium
as the oxide. The most imaginative
application
of
‘k
in medicine
is that
of a power
supply
for a
mechanical
pump
to totally replace the human heart (Fig.
4).
In
this
case,
each
mechanical
heart
would
require
about
54 g, of ~
as ‘8Pu1602.
The reason,
of course,
for using the
160 oxide
in such applications
is to lower
the neutron
exposure
of the recipient
by eliminating
the
IXX reaction
that
occurs
&h
normal
abundance
170. If
the formidable
biological,
medical
and mechmicd
prob-
Iems
of
this
application
can
be
overcome
in
the
1990-2000
period,
Dr. Seaborg visualizes that there might
be as much as 6000 kg of ‘%% committed
to this use by
the turn of the century.

Env-ixonmen@l Plutonium
Contamination

Animal
experiments
beginning
with
the fwst injec-
tions
of
plutonium
into
rats
in
April
1944
by J. G.
Hamilton
and
his
colleagues
at
the
University
of
California,
have shown
unequivocally
that
this material,
taken
into
the body
in sufficient
quantity,
will produce
undesirable
effects
(including
cancer)
in anirrlals and un-
doubtedly
in man. If the role of plutonium
in our future
economy
is to approach
remotely
the projected
levels,

i

IHIELD
1

there must
be a continuing
program
to prevent
unaccept-
able buildup
of contamination
in the environment.
Gener-
alized
contamination,
as seems
to have
occur-red
with
mercury,
must
not
be allowed
to happen.
That
is why
professionals
such as you
attending
this symposium
are
important
now and
will become
progressively
more
im-
portant
in the future.
One can visualize a number
of ways
whereby
plutonium
may
be
discharged
advertently
or
inadvertently
into
the environment.
Potentially
at least,
nuclear
power
plants
can
disperse
plutonium
into
the
environment
through
improper
discharge
of gaseous and
liquid
effluents
and
through
accidents
that
disrupt
the
integrity
of containment.
Plutonium
processing
and fabr-
ication plants
can contaminate
the entiroment
through
improper
gaseous,
Iiquid,
and
solid
waste
management
and can have accidents
such as facility
fires and storage
and transportation
mishaps
involving the raw materials
as
well as the processed
or finished products.
Plutotium-238
thermoelectric
generators
can be involved
in fabrication,
transportation,
and deployment
accidents.
As exmples,
space power
generators
could
be involved
in launch-pad
explosions,
launch
abork
and orbit~
decay with reent~
and
atmospheric
bumup
or
impact
disruption.
(!on-
tarninated
waste management,
of course,
is of paramount
importance
in controlling
environmental
contamination.

6

Plutonium-238
heart
pacer

Fig. 3

Any
one
of
these
potential
sources
of environmental
contamination
could
constitute
an
entire
symposium
within
itself. I have purposefully
refrained
from mention-
ing
nuclear
weapons
and
weapons
testing
as potential
sources
of
environmental
plutonium
contamination
which, hopefully,
will disappear
in the near future.
In all cases, prevention
of environmental
contamina-
tion must rely on sound,
effective
engineering,
the effec-
tiveness
of which
must
be under
continual
surveillance
with appropriate
and practical
methods
of monitoring
and
analysis
which,
of course,
is the
primary
topic
of this
symposium.

Environmental
Plutonium
Contamination
in Relation
to
Man

Plutonium
released
to
the
environment
can enter
man either
directly
through
inhalation
of atmospherical-
ly-suspended
material
or indirectly
through
incorporation
into his food chain.

Atmospheric
Suspension
and
Inhalation.
Figure
5
shows
a schematic
representation
of direct
exposure
of
man
via inhalation
of atmospherically-suspended
pluto-
nium.
There
are two modes
of exposure,
the first being

7

.
.

Fig. 4

inhalation
of particles
from
the primary
contaminating
source prior to surface
deposition
and the second
inhala-
tion of particles
resuspended
in the atmosphere
from the
contaminated
surface
subsequent
to
deposition.
In the
first case, the material
to which the subject
is exposed
is
already
suspended
[that
is, the suspension
factor
(Sf) is
unity] . Conceptually,
at
least,
estimation
of exposure
under
this condition
is easier than
for the second,
since
exposure
is dependent
on air concentration
at the point
of interest,
particle
size distribution,
inhalation
rate, time
of exposure,
and
chemicrd
form
of the
plutonium.
Of
coume,
if
one
wishes
to
relate
exposure
back
to
the
primary
source
term
(e,g.,
discharge
from
a processing
plant stack, noncritical
detonation
of a plutonium-bearing
nuclear warhead,
etc.), the problem
is far more complex.
The problem
now requires
consideration
of a long list of
additional
variables
involving
meteorological
factors
and
physical aspects of the specific incident.
The second mode
of exposure,
inhalation
of resuspended
material,
is com-
plicated
even further
by introduction
of even more vari-
ables,
some
of which
are poorly
defined
if at all. This
mode
of exposure
is represented
on the right of Fig. 5.
The problem
now is to estimate
inhalation
exposure
of an

individual
living in a contaminated
area for a life time or
any fraction
thereof.
Undoubtedly,
exposure
will depend
on how much of the source term (in this case, the amount
of plutonium
deposited
on the surface)
gets resuspended
into
the breathing
zone
[that
is, the resuspension
factor
(Rf)] . Rf is dependent
on a staggering
number
of inter-
related
variables
involving
ill-defined
phenomena
that
within
themselves
vary from place to place and with time.
Among these are nature
of the contaminated
surface (soil
type,
vegetative
cover,
asphalt,
etc.)
and
local
micro-
meteorology
(turbulence,
wind velocity,
rainfall,
etc.). In
addition,
the fraction
of the source term (amount
deposit-
ed)
available
for
resuspension
varies with
time at some
rate interrelated
to such other factors as soil type, vegeta-
tive cover,
rainfall,
etc.
This
attenuation
of the source
term is designated
as ~
in Fig. 5 and has been estimated
at
-40
days
for
prevailing
conditions
at
the
AEC’S
Nevada
Test
Site.
In case these
are not
complications
enough,
still
another
is the
amount
of local
physical
activity
(vehicular
traffic,
grazing cattle,
plowing,
etc.) in
the
area
which,
incidentally,
will also perturb
Ap. At
present
at
least,
it
is virtually
impossible
to calculate
exposure
in
this
situation
from
first
principles.
This

.

8

4

1

Rf ==

,a2_
,0-8

Fig. 5

impossible
situation
led me and a former colleague (Dr. P.
basis of these resuspension
factors,
it was estimated
that
S. Harris)
to derive a resuspension
factor
empirically.
In
1956, under
the pressures
of a sudden
anxiety
over the
hazards
of
noncritical
detonations
of
plutonium-
containing
nuclear
warheads,
we performed
a series of
quick experiments
in an area of known su~ace
plutonium
deposition
at the Nevada Test Site. Air concentration
and
surface
deposition
measurements
had
been made
at the
time the contaminating
event occurred.
At two different
times after the event, air samplers
were set up and resus-
pended
plutonium
resulting
from extensive vehicular
traf-
fic in the area was measured.
From this we concluded
that
a resuspension
factor

Air Concentration
(in pg plutonium/m3
j =7xl@m-1
‘~
Surface Deposition
Plutonium
(in ug/m
)
.

applied
to
disturbed
Nevada
desert
conditions
and that
.
the
attenuation
factor
>s
35 days.
An
attempt
was
made
also
to
calculate
resuspension
factors
by
other
means
that
might
apply
under other
conditions.
Deriva-
tion
of a resuspension
factor
from
equilibrium
calcula-
tions with dusty
rural air gave a value of 7 x 1@.
On the

the life-time
tolera-ce
surface
dep&ition
levels for pluto-
nium were 0.7 #Ci/m2
and 7.0 #Ci/m2,
for the respective
sets of conditions.
On the basis of data collected
during
Nevada
Test
Operations
Plumbbob
and
Roller
Coaster
(during
which
resuspension
was studied),
the
life-time
tolerance
surface
concentration
was
estimated
to
be
70 pCi/m2
for
undisturbed
regions
comparable
to
the
Nevada desert.
My perpetration
and application
of the resuspension
factor
have added
more to my infamy
tharr all the other
infamous
deeds of a 26-yr career. In the first place, from
the
scientitlc
point
of view,
the resuspension
factor
as
presented
here
is aesthetically
nauseating
and
simple-
minded.
It assumes that the surface deposition
level in the
imrnediat e vicinity
is the all-important
factor in determin-
ing the air concentration
above the contaminated
surface
and ignores the myriad
of factors
on which
resuspension
depends.
In the second
place, the resuspension
factor
as
an empirically
derived value applies only to the conditions
prevailing
at the time of derivation.
Reported
values range
all the way from
about
10-2 to 10-11. Intuitively,
I feel

9

that a factor of about
10+ is a reasonable
average value to
use in estimating
the potential
hazard
of occupancy
of a
plutonium-contaminated
area; however,
intuition
is not a
convincing
argument.
This aspect
of the potential
rela-
tionship
of man to plutonium
environmental
contamina-
tion has been emphasized
primarily
to emphasize
the need
for much
more
very difficult
and sophisticated
work on
the resuspension
problem.

Plutonium
Incorporation
rnto
the
Food
Chain.
Figure
6 is a schematic
representation
of the steps along
the food chain from soils to man. Approximately
50% of
man’s food is derived from animal products,
according
to
the progression
on the left, and about
50% directly
from
plants,
according
to
the
progression
on the
right.
The
amount
of environmental
plutonium
transferred
to man
depends
on the
degree
to which
plutonium
is concen-
trated
or discriminated
against at each step in the progres-
sion. The ratio of the concentration
in the product
to that
in its precursor
is expressed
as a discrimination
factor. As

an example,
the
concentration
of plutonium
(taken
in
through
the root system)
per g of plant to the plutonium
concentration
per g of soil is about
5 x 10-s; however,
deposition
on plant
surfaces
may be a greater
source of
contamination
of plants
than uptake
via the root system.
Multiplication
of
the
discrimination
factors
along
the
progression
gives
a crude
estimate
of the
relationship
between
environmental
plutonium
contamination
and
man
via
dietary
intake.
The
discrimination
factors,
of
course.
are in some cases ordv crude estirnat=”
howf

~
u
1
Icant
only

,
.
..-,
... ..ever,
tiey
are
good
enough
to
show
that
incorporation
of
enm ronrnental
plutonium
contamrnauon
tnto man vta the
.~ when
the environ-
mental
contammatmn
~eveis are completely
intolerable
~o~Additional
ecological
studies
and more
r=ment
0:
ecological
discrimination
factors
are
needed,
however,
to
provide
public
assurance
and
to
establish
unequivocally
that
important
factors
have not
been missed. Uptake
of plutonium
is influenced
by chem-
ical form, and absorption
from the gastrointestinal
tract is

.

Fig. 6

10

a factor of about
100 higher ‘for very young animals than
for older ones of the same species. Also information
on
plutonium
uptake
and transmission
in aquatic
chains is
sparse indeed.
Certain aquatic
lower species are known to
concentrate
plutonium
by factors of 3000 to 4000. Effect
of environmental
modification
and
aging of plutonium
deposits
on
ecological
incorporation
should
be
con-
sidered.
All of these
considerations
require
continual
re-
finement
of monitoring
and analytical
methods
and the
development
of new techniques.
As you are al aware, one
of the most critical problem areas is that of representative
environmental
sampling.

In summary,
the projections
of plutonium
produc-
tion and utilization
during the next three decades are a bit
staggering to say the least. The technology
to produce
the
projected
amounts
is virtually
assured. Whether
the pro-
jections
offered
by Dr. Seaborg
and the Federal
Power
Commission
come
about
wilI depend
on
sophisticated
cost-effective
engineering
to control
environmental
con-
tamination
and
continual
entironmentaf
surveillance
to
check
on engineering
effectiveness
and
to convince
an
apprehensive
and
occasionally
skeptical
public
that
the
gain is worth the risk.

11

.

b

.

WORLDWIDE
PLUTONIUM
FALLOUT
FROM
WEAPONS
TESTS

by

John H. Harley

Health and Safety Laboratory

U.S. Atomic
Energy Commission

New York, N. Y.

ABSTRACT

The testing of nuclear weapons up to the beginning of the moratorium

distributed
about
300 kCi
of
‘?%
over the surface of the earth. Tests by

France and Communist
China have probably
added about 5% to that.

The concentrations of plutonium
have been measured in the stratosphere

and surface air. Over the past 10 years, data on deposition rate and cumulative

deposit are very scarce and information
on the plutonium
in the biosphere is

even scarcer.

The
introduction
of
17 kCi
of
‘6Pu
from
a SNAP
generator has in-

creased our
interest in the fate of
plutonium.
Additional
measurements are
being carried out
and
the
Health
and
Safety
Laboratory
has performed
a
ZMpu. Comwrable
data on
worldwide
soil sampling to evaluate distribution
of

‘9Pu
will also be obtained.

Plutonium
has been
produced
in both
the fission
and fusion
weapons
that
have been
tested.
The yield of
plutonium
per megaton
of explosive force varies consider-
ably as a function
of weapons
design but it is probably
valid to look at the total weapons debris and consider that
there is some average plutonium
yield.
Our work at the
Health
and Safety
Laboratory
or the work available for
discussion
is not
aimed at weapons
diagnosis and we are
largely confined
to considering
the ratio of ‘%
to %lr
as our yield indicator.
In this paper I will try to show how
much
plutonium
has been produced
in weapons
testing
and what the present distribution
is. The ‘%
introduced
by the
burn-up
of a SNAP-9A
device
is not strictly
a
matter
of weapons testing but it is certainly
related to the
overall
plutonium
problem
and
I will include
it in the
discussion.

Production
of’%

The
combined
testing
of
all the
nuclear
powers
througl
1962 had a fission yield of 200 Mt. This can be
translated
into
a production
of 20 MCi of %k
and
a
plutonium
production
of about 0.4 MCi. This latter figure
will be refined somewhat
in a later discussion.

Testing
in the atmosphere
during
the moratorium
has continued,
with France carrying out a number
of tests
in the southern
hemisphere
and Mairdand China a number
in the
northern
hemisphere.
The
total
fission
yield
of
these tests through
1969 has been about
5% of the yield
of the pre-1963
testing.’

Plutonium-239
Data

The
plutonium
data
that
are
available
include
measurements
in the stratosphere
with
aircraft
and bal-
loons,
measurements
of
surface
air,
measurements
of
monthly
deposition
rate,
and cumulative
deposition
on
the
ground.
Some
information
has been
published
on
plutonium
in the biosphere.
I will try to review the data
and
to
point
out
some
of the
inferences
that
may be
drawn.

Stmtosphere.
Measurements
of 23%
in the stratos-
phere
have been
part
of all the programs
in this region,
and
data
are available
from
1957 to date.
The balloon
concentrations
and ratios are not shown specifically
with
the aircraft
measurements
but are included
in the inven-
tories given later.

13

Concentrations
of ‘%% in the stratosphere
change
with
time,
with testing, and with meteorological
factors.
‘llms
I have
chosen
to
tabulate
the
‘?u/%%
ratios,
which
will change only
if the pattern
of weapon
types
changes.
Data
for the
High-Altitude
Sampling
Program
(HASP),
the
Stardust
Program
of the
Defense
Atomic
Support
Agency, and the Airstream
Program of HASL are
shown in Table 1. These ratios are suftlciently
constant
so
that the megaton weapons whoti
debris enters the stratos-
phere can be considered
as a single source.
The stratospheric
material
is removed
with
a half-
life of about
1 year. This is illustrated
for the %r
inven-
tory
in Fig. 1, and since the 239/90
ratio
remains con-
stant,
the
‘%
is leaving the stratosphere
at the same
rate.

Surface Air. Because plutonium
has been considered
to be almost
exclusively
an inhalation
hazard,
measure-
ments
have
tended
to
emphasize
surface
air concentra-
tions.
The
stratospheric
239/90
mean
ratio
of
about
0.017
may
be compared
with
the
surface
air ratios
in
Table II. The early data are in good agreement,
but later
ratios for surface air seem to be higher.
The actual concentrations
of ‘k
in surface air are
given in Table III. The Soviet data appear to be low, and
no check is available since %% was not measured.
Other-
wise, you might say that the mean level was about 0.1 PCi
per 1000 standard
cubic meters
for the years since 1965.
This is about
a factor
of
10s
below
the ICRP recom-
mendation
for
the
occupational
exposure
to insoluble
plutonium
at 168 h per week (10-11 #Ci/cm3 ). For solu-
ble plutonium,
the recommended
level is 15 times lower.
There were relatively
few measurement
of ‘k
in
surface air before
1965, but we can make some estimates
based
on
%r
data.
The
peak
concentration
in
the

~4

A
TOTAL
STRATOSPHERE

G
NORTHERNHEMISPHERE

G
3CUTHERNHEMISPHERE

t
lJiftGE ATMOSPHERICTESTS

~3

,;.

“3
b.

U)z :
L_d_d_d

1963”
la64
19S5
t%s
196T
196s
1969
1970

I&. 1

northern
hemisphere
occurred
in 1963, when a value of
100 pCi of %% ~er
1000 m3 was found.
This would be
about
2 pCi of
~u/ 1000 m3. On a broader
basis,2 the
no rthem
hemisphere
average
for
1963
was
about
40 pCi/1000
m3 for
%Sr and
the
average for
1958-59,
after
the
large
tests,
was about
10 pCi/ 1000 m3. These
would
correspond
to
0.8
and
0.2 pCi
‘%/1000
m3,
respectively.
The former value is in reasonable
agreement
with the values measured at IspraS and shown in Fig. 2.

TABLE I

239/90
RATIOS IN THE STRATOSPHERE

Program

HASP

Stardust

Airstream

Period
No. of Samples
239/90

8/57 - 6/60
342
0.017

6/61 - 12/61
13
0.019

1962
70
0.015

1963
44
0.016

1964
42
0.017
1965
182
0.017
1966
255
0.018

1967
207
0.021

1968
233
0.021
1969
209
0.016
1/70 - 8/70
160
0.017

Reference

(17)

(18)

.

.

(19)

14

103

102
~

E

Go

10’

Location

Winchester, Mass.

Over Atlantic

Japan

Ispra, Italy

Northern
Hemisphere

Southern
Hemisphere

TABLE II

239/90
RATIOS IS SURFACE
AIR

Period

5/65 - 2/68
3/68 - 3/69

67-68

58-66
67-68

7/61 - 12/65
.,. ,/
lYOO
1967
1968
1969
1970

1965
1966
1967
1968
1969
1970

1965
1966
1967
1968
1969
1970

I
I

I
[
I
I
I
I
I
I
1

1961 1962 1963 1964 1965 1966 1967 1968 1969 1!

Fig. 2

239/90
Reference

0.017
(15)
0.028

0.013
(5)

0.016
(16)
0.023

0.022
(3,14)
0.021
0.022
0.032
0.024
0.018

0.017
0.026
0.019
0.030
0.026
0.022

0.018
0.035
0.037
0.017
0.012
0.046

(2)

(2)

Deposition.
The actual
deposition
rate and cumula-
tive deposit
of “%
has received little attention,
largely
because
it was considered
to be of little significance,
but
also because
of the
tedious
chemistry
and comparative
lack of alpha spectrometers.
Since the deposition
of %
was well documented,
the
use of a general
239/90
ratio
should
give a good
estimate
of the plutonium
deposition.
Figure 3 shows the
latitudinal
distribution
of ‘%
as of 1967$
and multiply-
ing the ordinates
by 0.017 should give the 23%
distribu-
tion. Comparable
exercises with deposition-rate
measure-
ments should also be valid for most of the time period of
fallout.
The
increased
239/90
ratio
after
1965
must
be
considered
for more recent data on rates, but the cumula-
tive deposit was over 98$Z0down by 1965 and later deposi-
tion
has little
effect.
It must
be remembered,
however,
that
the %r
is decaying
at a rate of 2?4% per year. This
means
that
if we
accept
a 239/90
ratio
of 0.017
at
production,
the ratio would now be 0.023 for the present
cumulative
deposit.
The
worldwide
depositon
of
%Sr has been
esti-
mated as about
12.8 MCi, with the rest of the ‘%r being
accounted
for by decay and local fallout
at the test sites.
The corresponding’%
would then be about
300 kCi.

15

TABLE 111

23%
IN SURFACE
AIR

Location
Period

Winchester

USSR

Southern
Hemisphere

Southern
Hemisphere

64-69

65-66

1965
1966
1967
1968
1969
1970

1965
1966
1967
1968
1969
1970

23%,pCi/1
000 m3

0.02 -0.5

0.005

0.12
0.16
0.06
0.11
0.08
0.12

0.10
0.15
0.06
0.02
0.03
0.08

Reference

(15)

(lo)

(2)

(2)

80
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

60

40

20

n
“90
70
50
30
10
0
10
30
50
NORTH
SOUTH

Fig. 3

Plutonium
in the
Oceans.
A number
of measure-
ments
of plutonium
in surface
ocean
water
have been
made and Bowen, et al.s
have also measured
concentra-
tions
at depths
greater
than
500 m. Pillai, et al.c found
concentrations
of 2 to 3 pCi/ 1000 liters in the Pacific and
Miyake
and
Sugimura
and
Bowen
et al.’
found
levels
somewhat
less than
1 pCi/ 1000 liters. If you consider the
plutonium
to be uniformly
mixed in the region above the

16

thermocline,
there should be about

70
90

10pCi/1000
liters to
be comparable
to the land deposition.
Measurements
by Bowen et al. of the 239/90
ratio
showed about
0.006
for depths down to 400 m and twice
that
for depths
greater
than 500 m. His interpretation
is
that the plutonium
acts as a sedimentary
particulate
or is
possibly moving by biologicrd sedimentation.

G

Pilla~ et al. indicated
that kelp concentrated
pluto-
nium by a factor
of 1000, and that shellfish and fish gave
concentration
factors of 200 and 3, respectively,
as com-
pared
to an equal weight of sea water. Measurements
by
Wong,et
al.8 were in general agreement
although
the data
are extremely
limited.
Wong,et al. indicate
that the high
concentrations
found
in sediments
might be returned
to
the environment
through
the action of bottom
feeders.

Plutonium
in the Biosphere. The data on plutonium
in the biosphere
measured by alpha spectrometry
tends to
be very limited.
Magno, et al.9 measured
air concentra-
tions and total diet as well as human lung and bone during
the
period
1965-66.
Dietary
intake
was
measured
as
7 x 10-3 pCi/day.
The existing
air concentrations
would
have given an intake of about one-third
of this assuming a
breathing
rate of 20 m3 /day.
The lung samples averaged
O.45 pCi/kg
and
the
bones
ranged
from
0.04
to
0.12 pCi/kg.
The
only
comparable
data
was developed
in the
USSR by SmorodintsevAet
al. 10~ey
measured
air con”

centrations
in
1965
and
1966
to
be
about
0.005 pCi/ 1000 m3
and
found
lung
concentrations
of
about
0.15 pCi/kg. Their air concentrations
are unexpect-
edly low and should not lead to the lung levels found.
Smorodintseva
and
coworkers
again
also checked
the pulmonary
lymph nodes and obtained
concentrations
about
50
times
higher
than
the
hrng.
This
had
been
pointed
out
in earlier
work
on occupational
exposures
and is not unexpected.

Plutonium
Anomaly.
The 239/90
ratios
in surface
air during
1968-69 exceeded
the ratios in the stratosphere
11The apparent
enrichment
of’%
at comparable
times.
is not readily explainable,
although
it appears to be real.
tt is hoped
that
the data presently
being collected
will
help to clarify the situation.

Plutoniurm238

Our interest
in the problem
of plutonium
distribu-
tion and deposition
was revived when a SNAP-9A device
burned
up over the Indian Ocean in April 1964. This unit
was fueled
with
‘%
and the faUout systems
described
above became very useful in evaluating
the distribution
of
this material.
The original satellite contained
17 kCi of ‘%.
The
SNAP debris was first detected
in balloon
samples taken
over Australia at about 33 km in August of 1964. Material
then appeared
at aircraft
altitudes
in the southern
hemis-
phere
in May 1965
and
in the
northern
hemisphere
in
December.
It finally
reached
the ground
in the southern
hemisphere
in the spring of 1965. The stratospheric
inven-
tory of SNAP ‘8Pu is shown in Fig. 4.12
A number of the high-altitude
filters were examined
by Holland13 at Trapelo/West
to see if they
could
esti-
mate the particle size. Radioaut ography indicated
that the
23%
average
size
was
about
10 mw,
although
these

03
J II
I
1
1
I
I

1964
1965
1966
1%7
I%a
1%9

Fig. 4

particles
might very well have been associated
with larger,
inert dust particles.
The concentrations
of 23~u in surface air have been
measured
since
1964,
but
some
of the
early
data
are
suspect.
Table
IV shows
the 238/239
ratios
from
1965
on. It must be remembered
that ‘8Pu was also formed in
weapons
tests,
and
a ratio
of 0.03
is what
might
be
considered
characteristic
of test debris.
The
1968 ratio of almost
2 in the southern
hemis-
phere
points
out the different
origin of the two isotopes
and
their
different
behavior.
The
‘8Pu
was distributed
mostly
in the southern
hemisphere,
and was introduced
after most of the ‘i~u
had already been deposited.
We attempted
to follow
the deposition
on a very
limited
scale
by
measuring
monthly
samples
from
Melbourne,
Australia,
and New York City. Over the next
few years problems were encountered
at both stations
and
a considerable
fraction
of the data had to be discarded.
The
only
continuing
reliable
measurements
came
from
Ispra
where
the
EURATOM
group
analyzed
monthly
s 14 he
station
isn’t too suitable
for
deposition
samples
I .
estimating
the worldwide
distribution
and we have there-
fore embarked
on a program
of analyzing
soils.
Our
soil
sampling
started
last
fall
and
the
data
should be available this fall. Samples were collected
on a
worldwide
basis
by
HASL
staff
and
by
cooperating
scientists
in many countries.
All samples were taken
to a
depth
of 30 cm to insure inclusion
of all the fallout.
We

hope
that analysis of these samples for %r
will indicate
sample
validity.
Plutonium-239
will also be measured,
as
well as”%.

17

TABLE IV

238/239
RATIOS IN SURPACE
AIR

Location

Northern
Hemisphere

Southern
Hemisphere

Period

1965*
1966
1967
1968
1969
1970

1956*
1966
1967
1968
1969
1970

*Part of year only.

If the SNAP ‘%J
were uniformly
distributed
on
the earth’s surface, the area concentration
would be about
70 dpm/m2.
The problem is compounded
by the fact that
weapons
debris has been found at soil depths greater than
15 cm,
so we had to set 30 cm as our sampling
depth.
Since an average soil will run about
400 kg/m2
to 30 cm
depth,
we expect
to find about 0.2 dprnlkg for the SNAP
‘EPu. This automatically
sets the sample size at 1 kg and
requires
a leaching
procedure.
We believe that
we have
sufficient
data
to indicate
that plutonium
from weapons
tests
and SNAP debris
can be acid+ xtracted
from kilo-
gram quantities
of soil. This may not be true for samples
taken near the Nevada Test Site or even for all plutonium
processing
plants.
This would
have to be tested
on the
appropriate
samples.
There is some complication
in looking at plutonium
data in any sample.
The weapons
debris plutonium
con-
tains a small amount
of ‘~u,
probably
of the order of
3%. This value is not well established
because good alpha
spectrometry
was not being used on the samples that were
available to us in the pre-SNAP period. The ratio was not
a problem
during the time of major SNAP fallout because
238/239
ratios reached
1 and above. This will not be true
in the soil samples
to be analyzed
since we have only
17 kci
of 23Bpuplus about
half as much from test~gj
as

compared
to 300 kCi of’~
from the tests.

References

1. P. W. Krey and B. Krajewski, “Updating Stratospheric Inven-
tories to January 19707 USAEC Report HASL-239, January
1971.

2. “Results of Surface Air Analyses:’ USAECReport HASL-242
(Appendix), April 1971.

18

238/239
Reference

0.03
(2)
0.16
0.48
0.30
0.29
0.14

0.04,0.24
0.61
1.58
1.91
0.92
0.52

(2)

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

M. DeBortoli et al., “Pu-239 and 238, Sr-90, and CS-137in
Surface Air from Mid-1961-1965:’
IRPA Congress, Rome,
September 1966.

M. W. Meyer et al., “Strontium-90
on the Earth’s Surface,
IV:’ USAEC Report TID-24341.

V. T. Bowen, K. M. Wong, and V. E. Noshkin, “Plutonium-
239
in and
Over the
Atlantic
Ocean,”
USAEC
Report

NYO-2174-1 14.

K. C. Pfllai, R. C. Smith, and T. R. Folsom, “Plutonium in the
Marine Environment
Nature 203,568-70
(1964).

Y. Miyake and
Y. Sugimura,
“Plutonium
Content
in the
Western North Pacific Waters;’ Meteorological Research Insti-
tute (Japan), October 1968.

K. M. Wong et al., “Pu-239 in Some Marine Organisms and
Sediments;’
USAEC Report NYO-2174-1 15.

P. J. Magno, P. E. Kauffman, and B. Scfdeien, “Plutonium
in
Environmental
and
Biological Media;’
Health
Physics
13,
132s-30 (1967).

G. I. Smorodintseva
et aL, “Study
of Uptake of Airborne
Pu-239 by the Human Organism,” U. N. Scientific COmmittce
Document A/AC.82/G/L.1301,
HASL Translation, November
1969.

H. L. Volchok and P. W. Krey, “Plutonium-239
Anomaly in
the Troposphere;
USAEC Report
HASL-224,
1-14 to 27,
APrif 1970.

P. W. Krey, M. T. Kfeinman, and B. T. Krajewski, “Sr-90,
Zr-9S, and
Pu-238
Stratospheric
Inventories,
1967-1969:’
USAEC Report HASL-227, I-39 to 69, July 1970.

W. D. Holfand, “Final Report of Studies of Pu-238 Debris
Particles
from
the
SNAP-9A Satellite
Failure
of
1964;
Report TLW-6006, May 1968.

.

.

d.

14. Euratom
Joint
Nuclear
Research
Center,
Site Survey and
17. J. P. Friend, “The High Altitude Sampling Program;’
DASA
Meteorolofw
Section
Quarterly
Reports,
Reprinted
in the
Report 1300, Volume 3, August 1961.
HASL Qu&-terly Reports.
-
18. H. W. Feely, D. Katzman, and C. S. Tuc$k, “Sixteenth
Prog-
15. B. Schleien, 3. A. Cochran, and P. J. Magno, “Sr-90, Sr419,
ress Report on Project Stardust;’
DASA Report lfJ21.
Pu-239, and
Pu-238 Concen@ations
in Ground-Level
Air,
1964-1969/’ Envir. Sci. and Tech. 4,598402
(1970).
19. P. W. Krey and M. Kleinman, “Project
Airstream;’
USAEC
Reports, HASL Quarterlies, 1967-1971.
16. Y. Miyake, Y. Katsuragi, and Y. SugirnUra, “A Study
on
Plutonium Fallout,”
J. Geophysical Res. 752329-30
(1970).

19

I
I

I

I

I

1

DISTRIBUTION
OF PLUTONIUM

FROM
ACCIDENTS
AND
FIELD
EXPERIMENTS

by

Harry S. Jordan

Los Alamos Scientific Laboratory

University of California

Los Alamos, New Mexico

ABSTRACT

Studies of
plutonium
in the environment
from
accidents involving nu-

clear weapons and from
experiments
in the field
to study health and safety

aspects of operational
weapons are worthy
of
careful evaluation. Plutonium

fallout from
weapon testing is diminishing
and, for the immediate future, the

signing of the
Limited
Test Ban Treaty would
indicate that additions to the
inventory
will
only
be caused by
testing at a reduced rate by nations not

signing the treaty.
Plutonium
from
routine operations of plutonium
facilities
has never
been
a serious
problem,
and
the
current
AEC
drive
to
reduce

plutonium
contaminated effluent
to the lowest practical concentration should

reduce this source of plutonium
to a negligible level.

However,
as long as plutonium
exists as a component
of weapons, as

sources of
power
in
space as well as on
the
ground,
as a raw material in

fabrication
plants, and
as a waste product
in waste-handling
facilities, the

probability
of an accidental releaseof plutonium
to the environment
can never

be zero.

Reasons for
the
necessity of
desirability
to
study
the documented

accidents and field experiments are advanced and outlines of the accidents at

Thule,
Greenland
and Palomares, Spain together with the field exlx?riments,

Project 56, Project 57, and Roller Coaster are presented.

..-... —-------------------------------------

Nowadays,
discussions
regarding plutonium
seem to
have
a
certain
element
of
unreality
associated
with
them -- perhaps characterized
somewhat
by the expression
“The Wonderful
World of Plutonium.”
There
is even in
some cases a reluctance
to enter
into such discussions,
as
if it were rather
like talking about the virtues of marriage
in front
of your old maid aunt. There is really no reason
for this because
plutonium,
as a metal,
has a fine and
exceptional
history.
By that
I mean
that
materials
are
used by humans
in their affairs for good or for ill, but in
the course of this service the materials evolve a history
of
their
own.
Almost
all the common
metals and materials
such as coal and cotton
have long fascinating
histories
in
which the bright chapters
are blight ed by very dark chap-
ters. Plutonium,
in comparison,
does not, and should not

in the future,
have such blots upon its history.
We have
had almost
30 years of documented
experience
to indi-
cate that our present
knowledge
and techniques
are suffi-
cient
to
handle
this
material
in quantity
with
a real
margin
of
safety.
It is perhaps
worth
noting
that
the
accounting
for
illness,
death,
and
misery
that
can be
attributed
to
other
metals
and
toxic
materials
is very
incomplete
and fragmentary,
whereas
rather
careful
sur-
veillance
of the people
working
or involved
with
pluto-
nium has established
its remarkable
safety record.
In the years that we have been using plutonium,
it
has found its way into our environment
by three principle
means. The source that accounts
for the most widespread
distribution
of plutonium
is that created
in the upper air
by atmospheric
testing of nuclear weapons.
This source of

21

plutonium
has been diminishing
since the signing of the
Limited Nuclear Test Ban Treat yin
1963.1
Plutonium,
in small amounts,
has been
dispersed
into local environments
by effluents
from facilities hand-
ling plutonium,
but this dispersal
has been carefully
con-
trolled
and has not created a health hazard. Moreover, the
current
well-financed
AEC effort
to
reduce
plutonium
concentrations
in effluents
to the lowest
practical
level
will, for all practical
purposes,
eliminate
any real concern
about plutonium
in the environment
from this source.
Plutonium
dispersed
into
the environment
as a re-
sult of accidents,
however,
will always, in some measure,
be a problem.
If we are to make full use of this metal as a
vital element
in our natioml
defense efforts,
as a power
source
in space,
as well as on land and sea, and as an
element
in medical devices, we must accept
the certainty
that
accidents
will happen
and
that
plutonium
will be
distributed
to some extent
in our environment.
A large portion
of the information
that
has been
developed
concerning
the
dispersal
of plutonium
from
accidents
is in classified
documents.
Certainly,
access to
classified
information
is required
to
completely
under-
stand
the
reports
of the
actual
accidents
and the field
experiments.
I was going to say that this is unfortunately
the
case,
but
in reality
it is fortunate
that
accidents,
except
in the case of nuclear weapons,
have not created
any major environment
al health problems.
Probably
the first release of plutonium
was an ex-
periment
conducted
by the IKSSAlamos Scientific
Labora-
tory
in
the
early
days
of
the
Nevada
Test
Site.
The
purpose
of the experiment
was to determine
the proper-
ties of plutonium
when subjected
to forces generated
by
the detonation
of high explosives.
The monitoring
effort
was directed
primarily
toward
protection
of the workers.
This
event
is
mentioned
here
only
to
note
that
AEC-NVOO
has appointed
a committee
to
study
sites
with old plutonium
contamimtion
and that the Reynolds
Elect rical and Engineering Company,
the support
contrac-
tor for the Nevada Test Site, is now engaged in collecting
preliminary
data from this area.2
The
first
field experiments
for evaluating
weapon
safety
were
conducted
by
the
Los
Alamos
Scientific
Laboratory
in
1955
and
1956
in an operation
called
Project
56. These experiments
were required
to establish
design
parameters
to
ensure
that
weapons
involved
in
accidents
would
not
produce
a nuclear yield. A total of
four events was necessary to develop the needed data. The
study of the plutonium
contamination
levels produced
in
the environment
by the experiments
was considered
to be
of secondary
importance,
but a quickly
assembled
group
of people produced
data on air and ground contamination
levels as a function
of distance. 3 As part
of the overall
effort,
persomel
on
the
H-Division
staff
of
the
Los
Alamos Scientific
Laboratory
produced,
on a crash basis,
a hazard
evaluation
for the release of plutonium
from a
weapn
involved in an accident.4
One of the conclusions
that evolved from this theoretical
evaluation,
bolstered
by
scant
environmental
data
from
Project
56,
was
that

100 #g Pu/m2
on the ground
would be safe for a lifetime
occupancy.
The authors,
well aware of the uncertainties
and assumptions
that had gone into this urgently
needed
evaluation,
strongly
recommended
additional
studies
of
the accident
case.
The need for additional
data, acknowledged
by the
AEC and DoD, led to the experiment
conducted
by Test
Group
57 Operation
Plumbbobs
in
1957.
Four
broad
areas of interest were studied.

.
Means of estimating
&Mbutkmand long-term
redistribution
of
plutonium
dispersed
by
a nonnuclear
detonation.

G Biomedical
evaluation
of a plutonium-laden
en-
vironment.

G Evaluation
of decontamination
methods.

.
Alpha survey instrumentation
and field monitor-
ing procedures
for the
prompt
estimation
of levels of
plutonium
contamination.

The
various
studies
produced
data
that
should
be
more widely distributed
and should be subjected
to addi-
tional
analysis.
The figure of
35oO pg Pu/m2
was estab-

lished as safe for lifetime occupancy
with normal activity,
i.e.,
weather
as the
sole
resuspension
force,
being
an
important
stipulation.
The number
generally
associated
with Project 57, however, is 1000 pg Pu/m2.
Data from this single release did not settle all of the
questions
and uncertainties
that
bothered
the AEC and
DoD in their
efforts
to develop
proper
criteria
for the
storage
and
transportation
of nuclear
weapons.
Sharing
this
concern
was the
Atomic
Energy
Authority
of the
United Kingdom.
The three agencies therefore
sponsored
Operation
Roller Coaster.
The field experiments
were carried
out jointly
by
United
States and United
Kingdom
personnel
in 1963 on
the Tonapah
Test Range in Nevada. The objectives
of the
operation
were:

.
TO make measurements
of plutonium
to permit
its distribution
and
behavior
during
cloud
travel
to be
determined.

.
TO obtain
data
to permit
complete
characteriza-
tion of the aerosols
in the cloud.

.
TO determke
the
lung deposition
and
fate
of
plutonium
inhaled during cloud passage by several animal
species;
to
compare
animal
data
with
air sampling
data
and attempt
to estimate
the dose to man from inhalation.

.
TO evaluate
the effects on the dispersal of pluto-
nium of varying
amounts
of earth
caver on storage con-
figurations.

22

.

.
TO further
develop
the model
describing
cloud
behavior
and particle deposition
using sedimentation
and
turbulent
diffusion
theory
so that plutonium
releases in a
variety of weather
conditions
could be estimated.

Prior to conducting
the experiment,
United
States
and
United
Kingdom
personnel
had
agreed that
no at-
tempt
would
be made
to obtain
resuspension
measure-
ments
because of the complex
nature
of the process and
the effort
required.
The difficulties
and complexities
are
not to be denied, but the inability
to fund or to interest
a
qualified
group
to investigate
the resuspension
of pluto-
nium is a matter
to be regretted.
This is particularly
true
since much
of the basic data
required
in a resuspension
study,
i.e., the level of plutonium
contamination
on the
soil, was established
at great cost and effort by the various
test groups involved in this operat ion.
Altogether,
four
experimental
field
releases
from
four
shots
were
involved
in the
Operation.
Two
shots
were conducted
in the open with a difference
in the ratio
of plutonium
to high explosive, and the other two were in
a s~orage configuration
with the same ratio of plutonium
to high explosive
but with
a difference
in the depth
of
earth overburden
on the storage structures.
The
experiment al arrays
were
elaborately
instru-
mented
for
the
detection
of
airborne
plutonium
and
deposited
plutonium.
A heavily
instrumented
wire cur-
tain,
1500 ft in width
and for one shot 1800 ft in height,
was used to document
the vertical distribution
of pluto-
nium in the cloud.
On one of the events a total of 298
animals (84 beagle dogs, 84 burros,
and 130 sheep) were
positioned
in the downwind
instrumented
array.
These elaborate
field experiments
developed
a great
mass of data and resulted
in a large number
of published
reportsG>7 on the
various
projects.
It is clear, however,
that
additioml
efforts
should
be devoted
to an analysis
and correlation
of this data.
It should
be noted
that
over the intervening
years
the
Reynolds
Electrical
and
Engineering
Company
has
periodically
resurveyed
the areas that were contaminated
by these experiments.
The wisdom
and foresight
of the authorities
who
decided
to
conduct
the
field
release
experiments
were
validated
by the Palomares
and Thule
accidents.
In the
first place, and of the utmost
importance,
the bombs that
did explode
as a result
of the accidents
did not give a
nuclear yield. Secondly,
the experiments
created
a group
of people within
the AEC and DoD communities
with an
understanding
of,
and
a thought-out
position
on,
the
problem.
Fortunately,
one member
of this group,
Wright
Langham,
Los Alamos Scientific
Laboratory,
was brought
in as a DoD consultant
on both accidents.
It was primarily
through
his efforts
that this country
was able to arrange
satisfactory
agreements
with both the Spanish and Danish
authorities.
Basically,
the
operative
procedure
was
to
make available to the Spanish
and Danish authorities
the
resources
required
for
them
to
assure
themselves,
and
consequent ly their
people,
that
the
hazards
had
been

properly
evaluated
and eliminated.
Published
papers
by
persomel
of both
countries
have indicated
that
such is
indeed the case.s-ll
The
Palomares
accident
on January
16, 1966, re-
sulted
from a mid-air explosion
during a refueling
opera-
tion between
a B52
bomber
and a KC-135 tanker.
Four
plutonium-bearing
nuclear
weapons
were
jarred
loose
from
the
plane
by the explosion.
Three of the devices
impacted
on the
ground
in the vicinity
of the Spanish
village of Palomares
and one landed
in the Mediterranean
Sea. Two weapons
were ultimately
recovered
intact,
the
one from
the sea and one of the three that impacted
on
the ground.
The other
two weapons detonated
on impact
with
the
ground
and
dispersed
plutonium
over
some
1200 acres of ground.
A wind with an estimated
velocity
of 30 knots
prevailed
at the time. It should be noted that
under
these
conditions
the radius of the area with con-
tamination
over 500 #g/m2
was about 80 m for one deto-
nation
site and about 65 m for the other site. The cleanup
procedure
consisted
of scraping
and
removing
the
top
layer of soil from about 6 acres with contamination
levels
above
500 gg/m2.
Crops,
in fields
with
contamination
levels above 5 ~g}mz, were removed and de&oyed.
All of
this material
was packaged
and ultimately
shipped
to the
United
States.
Originally,
it was planned to plow only the
land
between
the
50 #g/m2
and 500 pg/m2
contamina-
tion contours,
However,
with equipment
on hand, it was
decided
to plow to a depth
of about
10 in. all the land
contaminated
to a level above 5 Ug/m2. It was considered
that the plowing would dilute the plutonium
by mixing it
with a greater mass of soil and would make the plutonium
less available
for resuspension.
As previously
noted,
the
Spanish
authorities
have reported
that after the area was
decontaminated
the
air
concentrations
in
the
vicinity
were those
to be expected
from
worldwide
fallout
and
that all determinations
for plutonium
uptake
on the part
of the inhabitants
of Palomares
had been negative.
The
crash of a B-52 bomber
on the ice of North
Star Bay about
7 to 7% mi from
Thule,
Greenland,
oc-
curred
on January
21,
1968.
Cause of the crash was an
uncontrollable
onboard
fire that made it necessary for the
crew to bail out.
The plane impacted
on the ice with a
velocity
of about
500 knots
and at a 15 degree attitude.
On
impact,
the
fuel
ignited
and
the
four
plutonium-
bearing
weapons
exploded.
Debris and flaming fuel, pro-
pelled by the forward
motion
of the plane, was scattered
along a path
about
700 m long. A large blackened
area
about
130 m wide and 700 m long was formed
by com-
bustion
products
being trapped
in refrozen
ice and snow.
It
has
been
estimated
that
approximately
99% of the
plutonium
within
the
defined
contaminated
zone
was
contained
in the black crusted
ice and snow of this area.
Road graders windrowed
the black material
and mechan-
ized loaders placed it in large wooden
boxes for removal
from
the
contaminated
area.
Eventually
sixty-seven
25,000-gallon
fuel containers
were ffled
with this mate-
rial and four additional
such containers
were required
to
store contaminated
equipment
and gear. This material was
shipped to the United States for final disposal.

23

A cloud formed
by the explosion
was measured by
radar
as being about
850 m high, 800 m in length,
and
800 m in depth,
and it undoubtedly
carried some pluto-
nium downwind.
Danish scientists
investigated
rather
thoroughly
the
levels of the plutonium
in the environment
and concluded
from
their
findings
that
the
environmental
impact
was
negligible.
Two important
points
that
should be remembered
are demonstrated
by the experience
from these two acci-
dents.
First,
the
dispersal
of appreciable
quantities
of
plutonium
did not create a catastrophe
in terms of human
impairment
and
death
or in terms
of property
damage
but,
instead,
were
incidents
that,
with
modern
tech-
nology,
were
brought
under
rather
complete
control.
Secondly,
the determination
to assist the local authorities
in their evaluation
of the situation
made it possible for
them
to convince
themselves
that
humans
had not been
injured
by
the
immediate
effects
and
that
long-range
hazards
had
been
eliminated
or reduced
to acceptable
levels. This assurance
was conveyed
to their citizens and
appreciably
reduced
the strain
on our international
rela-
tions.
1 would sincerely
hope that our own citizens would
be treated
with the same consideration
and respect in the
event of a similar incident
on United
States soil. I have
instead,
however,
a very unhappy
vision of such an event,
in which
the
news
media
are on an anti-establishment
kick, security
and atomic
energy experts
indulge in indi-
vidual ego trips,
and credibilityy is completely
destroyed,
with the final result being a group of citizens unhurt
and
unendangered,
but
compelled
to
carry
a psychological
burden
of worry,
fear,
and doubt
for the rest of their
lives. That
may
be an unduly
pessimistic
vision, but
it
does
seem
clear that
positive
steps should
be taken
to
identify
the best possible response
to an accident
involv-
ing plutonium.
A suggested
first step would
be to fund
a serious
effort

.
To compile
and evaluate
available data from the
field releases and the accidents,

.
TO
provide
m
unclassified
arid realistic evaluation
of the
hazards
associated
with
an accidental
plutonium
release to the environment.

G To identify
those
areas that require
funding
for
immediate
and long-range investigations.

A realistic
evaluation
would
in large measure
offset
the
harm
that
has been
done
by the misapplication
of the
“maximum
credible accident”
concept,
and would help to
define
plutonium’s
proper
place
in
the
spectrum
of
hazards
that confront
man in a modern industrial
society.
If this is not accomplished,
and plutonium
is compelled
to
occupy
a unique
position
completely
outside
this spec-
trum,
then very likely the ultimate
judgment
will be that
science and technology
have again been mismanaged.
The
dissatisfaction
with
science
today
stems
basically
from

24

our
apparent
inability
to realize
the
benefits
of
tech-
nology
without
undue
impairment
to our physical,
envi-
ronmental,
and
social
well-being.
It
has
been
demon-
strated
that
the
benefits
of
plutonium
can be realized
with
minimum
adverse
impacts
on our society.
Forcing
plutonium
out of the marketplace
by unnecessary
restric-
tions
will
only
encourage
and
prolong
dependence
on
materials
that
have had
in the
past, and probably
will
continue
to have in the future,
severe detrimental
effects
on society.

References

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

James H. McBride, Z7re Test Ban ZYeaty, Hemy Rcgnery Co.,
Chicago, Illinois, (1967).

Plutonium
Environmental
Studies Program,
Reynolds
Elec-
trical and Engineering Co., report in preparation.

Witfiam S. Johnson,
“Report
of Fallout
Study
of January
1956, 56 Project
NTS~
Los Alamos Scientific
Laboratory
Report LAMS-2033 (June 1956), (Classified).

P. S. Harris, E. C. Anderson, and W. H. Langham, “Contamin-
ation
Hazard from
Accidental
Non-Critical
Detonation
of
Small Atomic
Devices:’
Los Alamos Scientific
Laboratory
Report LA-2079 (September 1956), (Classified).

J. D. Shreve, Jr., “Operation
Phsmbbob-Test
Group 57, Re-
port
of
Director
Test
Group
57:’
Sandia
Corporation,
Albuquerque,
NM,
Report
ITR-1515
(April
1958),
(Classified).

J. E. Shreve, Jr. and D. M. C. Thomas, “Operation
Roller
Coaster; A Joint Field Operation
of the Department
of De-
fense,
the
Atomic
Energy
Commission,
and
the
United
Kingdom Atomic Energy Authority (AWRE); Scientific Direc-
tor’s Summary Report~’ (Classitled).

K.
Stewart,
“Rotler
Coaster,
Summary
Report;’
United
Kingdom Atomic Energy Authority Report AWRE No. T6/69
(July 1969), (Classified).

Ernitfio Iranzo
and Sinesio Salvador, “hrhalation
Risks to
People Living Near a Contaminated
Area:’ Junta de Encrgia
Nuclear,
2nd
International
Congress
of the
International
Radiation
Protection
Association,
Brighton,
England, (May

3-8, 1970).

Eduardo
Ramos Rodriguez, “Prdomares - Two Years After:’
(APril 1968).

USAF Nuclear Safety, AFRP 122-1 Jan/Fcb/Mar/1970,
No.
1, Volume 65 (Part 2), Special Edition Project Crested Ice.

G
f) a nkh
Thule
Committee,
“Evaluation
of
Possible
Hazards:’ pp. 8-11.

G
Jorgcn Koch, “Danish Scientific Group [nvestigations~
pp. 4244.

G
Hemy
L.
Gjorup,
“Investigation
and
Evaluation
of
Contamination
Levets:
pp. 57-63.
.
Christian Vibe, “Ecological Background:
pp. 64-69.

G
F. Hermann, “Ecology Survey:’ pp. 70-73.

G
Asker
Aarkrog,
“Radio-Ecological
Investigations;’
pp.
74-79.

G
Walmod Larsen, “Danish
Health Physicists’ Activities:’
pp. 80-81.

A Aarkrog, 4’Radioecologicrd Investigations of Plutonium
in
an Artic Marine Environment:’
Heatth Physics, Vol. 20 (Jan.
1971), Permagon Press, pp. 31-47.

.

.

INDUSTRIAL-TYPE
OPERATIONS
AS A SOURCE
OF
ENVIRONMENTAL
PLUTONIUM

by

S. E. Hammond

The Dow Chemical Company

Rocky Flats Division

Golden, Colorado 80401

ABSTRACT

From
1953
through
1970,
the
Rocky
Flats plant
has releasad uppar

limits of 41 mCl of plutonium
as airborne effluents and 90 mCl of plutonium

through
liquid
eff Iuants. Methods
and limitations of these measurements are

described.
In addition
to these controlled
releases, accidental releases to tha en-

vironment
occurred
during
a fire
in
1857
and
from
wind
transference of

contaminated soil prior to 1970. Thasa incidents are described and estimates of
amounts of plutonium
involved made by various investigators discussed.

... .....-
---..
—.. -.-..
—.. —.-. .-.. -- ........-

The
ROCQ
Flats
plant
began
operating
in
1953
processing
plutonium,
enriched
uranium,
and
depleted
uranium.
Over the
years
more
and
more
emphasis
has
been placed on plutonium
and less and less on the other
mat erials.
Since
this
is a plutonium
meeting,
we will
confine
our discussion
to plutonium
operations
at Rocky
Flats.
Figure
1 is a map
of the
area
in which
we are
located.
The AEC-wned
land is 2 miles on a side with the
occupied
portion
of
the
plant
site confined
to
about
1 square
mile
in
the
area
between
Walnut
Creek
and
Woman Creek. Downtown
Denver is about
15 or 16 miles
to the southeast.
The sout hem
city limits of Boulder,
a
city of 70,000,
lie 6 miles north.
The other towns shown
are smaller.
This area
is essentially
greater
Denver
and
urban. The area close to the plant is rural -- mainly grazing
land although
there is some irrigated
farming.
Plans exist
for
commercial
and
residential
development
close
by,
mainly to the south and east.
The southern
portion
of the plant site is drained by
Woman
Creek,
dry
part
of the
year,
which
flows into
Standley
Lake. Standley
Lake is an irrigation
reservoir as
well as the municipal
water supply for Westminster.
The
northern
portion
of the plant is drained
by two branches
of Walnut
Creek which join east of the plant
and flow
into
Great Western
Reservoir.
Great Western Reservoir
is
Broomfield’s
municipal
water supply.
Effluents
from our
process
waste
treatment
plant
and
from
our
sanitary

sewer system flow into the south branch of Walnut Creek
and through
a series of four ponds before
release offsite.
Walnut
Creek
provides
about
2% of
Great
Western’s
water,
another
8% comes
from
Coal Creek,
and the re-
maining 90% from the Clear Creek watershed.
The foothills
of the Rocky Mountains
extend
along
the west edge of Fig. 1; the remaining
terrain is typically

ROCKYFLATSFtJWT

Fig. 1

25

prairie - arid
and sparsely
vegetated
except
where
it is
irrigated.
The government-owned
land is enclosed
with a
barbed
wire cattle
fence;
there
are no domestic
animals
within
its boundaries.
Wildlife which
shares
our domain
includes
such typical
prairie types as deer, coyote,
rattle-
snakes, and rabbits.
Winds
from
the
west,
northwest,
and
southwest
prevail
along
the
foothills.
During
the
fall and
winter
months
windstorms
occur
frequently.
Gusts
over
100 mph have been recorded.
The prevailing winds at the
Denver weather
station
are from the south.
Preoperational
site-survey
measu~ements
were con-
ducted
by a team from Hanford
and included
beta-gamma
surveys,
and water
and vegetation
samples
analyzed
for
uranium
plus plutonium
content,
plus a few radium meas-
urements.
Figure
2 is a close-up
of Rocky
Flats.
This figure
shows our original plutonium
processing facility, building
771,
and process
waste
treatment
facility,
building
774.
More
recent
additions
to
the
plant
include
buildings
776-777,
a production
building completed
in 1957, build-
ing 779 R and facilities
completed
in 1966, building 559,
an analytical
laboratory
completed
in 1968, and building
707,
a production
building
completed
in 1970. We refer
to this entire
area as the plutonium
complex.
Plutonium
operations
are confined
to this area for the most part. The

903 area was used as a temporary
storage
area for drums
containing
contaminated
oil for a time.
We will discuss
this more later.
All effluent
air from plutonium
buildings is filtered
through
HEPA falters and stacks are continuously
moni-
tored
for
airborne
releases.
Isokinetic
samples
are
col-
lected
through
HV-70
paper
and evaluations
are calcu-
lated in terms of total long-lived alpha. Initially, when we
believed
that
all releases
were of PU02,
we applied
the
guide
level of
1 pCi/m3
for insoluble
plutonium.
Now,
rather
than
demonstrate
proportions
of
insoluble
and
soluble plutonium
in typical effluents,
we apply the more
restrictive
soluble guide of 0.06 pCi/m3 for soluble pluto-
nium to all stack releases.
Figure
3 shows
the
location
of our
12 onsite
air
samplers.
These
are cent inuous
samplers
drawing
2 cfm
and are collect ed daily. Total long-lived alpha evaluations
of these
samples have always been well below the pluto-
nium guide
levels. Figure
4 shows typical data from the
onsite
sampling
net.
This particular
display
is for
1968
and 1969 and is no different
from earlier years.
We believe
that
additional
filtration,
advanced
de-
sign features,
and
more
exhaustive
treatment
of liquid
wastes,
some
already
completed
and
some
yet
to be
completed,
will pIace us in a position
of near zero release
within the next few years.

.

b

.

Fig. 2

26

ROCKY FLATS
PLANT

/7/
/
//
J

/a’”
I
‘-7 ‘ WI,-,,\
~i -J’

1
Kno
k’
=
ALL!!l!! ‘“

}11
II
L--
--==4

!di
pcldl
r—m--ll.cll
IL

I
Ill-
\

1968-1969 ON-SITEAIR SAMPLES-TOTALLNG LIVEOALPHA
MONTHLYAVERAGES

0.1
1

pCi/ms 001 :

Fig. 4

Gs.7
I

Fig. 3

/

Figure
5 shows average total,
long-lived alpha con-
centrations
measured
in the building
771
main exhaust.
This
particular
building
typically
shows
greatest
opera-
tional releases. The graph indicates
both yearly averages as
well as the range of monthly
averages in pCi/m3.
The high
point
in 1957 occurred
following
a fire that damaged
the
filter
system.
The
high
points
of
1964 and
1965 were
attributed
to falter leakage occurring
about
the middle of
December,
1964
and
corrected
in
the
latter
part
of
January,
1965.
Figure
6 shows total
stack release by year from our
plutonium
complex.
The
data
are expressed
as gCi
of
total
long-lived
alpha.
The
high
concentrations
seen in
Fig. 6 are the
1957 fire and building
771 filter failure
in
1964
and
1965. The 1957 peak does not represent
total
release during
the fire since our sampler
became
inopera-
tive during
the
fire.
Rather
it is an indication
of high
samples observed
in October,
1957 from contamination
in
the
ductwork
and
plenum
following
restoration
of the
system.
The peak in 1969 is due to higher samples
from
building
776 following
the May
1969 fire.
Figure
7 de-
picts
integrated
airborne
releases
through
the stacks and
totals
41.3 #Ci
of total,
long-lived
alpha
through
April
1971.
Waste
solutions
generated
in the plutonium
com-
plex include
laundry
wastes
and process waste solutions
generated
at
various
phases
of
the
operations.
Such

27

TOTAL
LONG-LIVED
ALPHA
CONCENTRATION
100.01,
1
#
I
1
1
I
#
#
u
1
I
I
I
1
8

~ BLDG 771 MAIN EXHAUST

10.0 -

I.0 -

pCi/m3

o. I

0.0 I~ II I I

GYEARLY
AVERAGE

I MONTHLY AVERAGE
RANGE

1

1

0.001’ ‘

1
1
,
1
1
1
1
1
I
,
1
1
a
1
1
n

53
54
55
5657
58
59
60
61
62
63
64
65
66
67
68
69
70

PWTONIUM STACK RELEASED
TOTAL LONG-LIVED ALPHA

I

I& ;

I&:

10 1953 55
57
59
61
63
65
67
69
71

Fig. 6

I

28

YEAR

Rg.5

solutions
are held
in storage
tanks
at their
generation
point
until the y have been analyzed,
at which time several
options
exist. Solutions
which are low in plutonium
con-
tent but high in chemical content
may be pumped
to solar
evaporation
ponds
for
concentration.
Solutions
which
meet
USPHS drinking-water
standards
in chemical
con-
tent and 10 CFR 20 standards
in radioactive
content
may
be released
to the sanitary
waste system.
Using the same
rationale
as with
airborne
effluents,
we apply
the most
restrictive
guide of our plant materials,
1600 pCi/liter for
soluble
plutonium,
as our release
point.
Other solutions
are pumped
to building 774, our waste treatment
facility,
for
further
plutonium
removal.
Solids
resulting
from
building
774
operations
and
other
solid
plutonium-
containing
wastes generated
in other
buildings
are pack-
aged and shipped
to Idaho for burial and storage.
There
have been
no known
plutonium
releases to the environ-
ment by way of solid waste handling.
Liquid
effluents
from
building
774 are released
to
the south
Walnut Creek
course when
they meet USPHS
and
10 CFR 20 guides.
This effluent
joins
with sanitary
sewage effluent
and flows through
a series of ponds into

&

A

50

40

30
mCi

Ic

f

INTEGRATED AIRBORNE LONG-LIVED ALPHA ACTIVITY

>
8
[
n
E
I
1
1
I
n
8
I
I
I
a
,
,
1

CONTROLLEDRELEASES THROUGH EXHAUST SYSTEMS

’53
55
57
59
61
63
65
67
69
71
YEAR

Fig. 7

Great
Western
Reservoir.
Figure
8 shows
ponds
1 on
north
Walnut Creek, 2through5
on southWalnut
Creek,
and
9 on
Woman
Creek.
Ponds
1, 5, and
9 serve as
monitoring
ponds.
Ponds
1 and 9aregrab-sampled
daily
and
samples
composite
for
a weekly
analysis.
Pond
5
outflow
is sampled
continuously
by a proportional
sam-
pler
and
analyzed
weekly.
Following
the
lead
of the
Hanford
preoperational
site survey team we performed
a
so-called gross alpha analysis on these samples for many
years.
The analysis
actually
is specific
for uranium
and
plutonium
and
separates
out
other
alpha emitters.
The
most
rest rictive
guide,
for
soluble
plutonium,
has been
applied
to the gross alpha activity.
Now we also analyze
these gross alpha samples by alpha spectrometry
to deter-
mine specific
plutonium
content
as well. Figure 9 shows
the gross
alpha
cent ent
of pond
5 effluent.
This value
includes
natural
uranium
found
in Colorado
waters.
The
upper
line is the maximum
sample
found
in a year, the
bottom
line the
yearly
average.
As a comparison
with
plutonium
concentration
our
1970
measurements
aver-
aged 2.8 pCi/liter
phrt onium with a maximum
single sam-
ple of 8.6 pCi/liter.
Figure
10 shows
the
integrated
amount
of gross
alpha
activity
released
through
pond
5 to be 88.5 mCi.

Prior
to
the
addition
of building
778,
the
plutonium
laundry
was located
in building
771. Laundry
waste sam-
ples lower than
1600 pCi/liter
were released
directly
to
the north
branch
of Walnut Creek. An additional
2.5 mCi
of activity
has been
released to the environment
by this
route, or a total of 91 mCi in liquid effluents.
An undeter-
mined portion
of this 91 mCi is naturally
occurring.
in addition
to
controlled
releases,
plutonium
has
been released to the environment
from thsee occurrences.
In September,
1957 plutonium
metal spontaneously
ignited
in a glovebox
and several kilograms
burned.
The
fire quickly
burned
though
the Plexiglas window.
After
unsuccessful
attempts
to control
the fire with C02 a fine
spray of water
was used successfully.
Large amounts
of
smoke
had
filled
the
room
and
the
exhaust
fans were
turned
on high speed to clear the smoke. This smoke was
plainly
visible as it left the stack. However,
portable
air
samplers set up to monitor
this smoke detected
very little
long-lived activity.
The
fire next
spread
up the exhaust
ducts
to the
exhaust
filter plenum.
Flammable
filters soon caught fire
destroying
a major
portion
of the filtering
system.
‘Ilk
spread
of fire was accompanied
by an explosion
in the
exhaust duct.

29

1“
#
. . .
,.
..

Fig. 8

30

URANIUMtl PLUTONIUMALPHAACTIVITYIN~D
5 EFFLUENT
100 ~.
Y

MAX

“’’”o
:~:
[......ll......ll

‘“0 54
56
58
60
62
64
66
68
70
YEAR

Fig. 9

Following the fire it was estimated
that about
1 g of
plutonium
had been released offsite through
the damaged
fflter system.
Figure
11 is the alpha spectrum
of a 15-min, high-
volume
air sample
taken
downwind
(south)
during
the
fire. It indicates
a concentration
of about
4 pCi/m3 Pu.
Another
sample taken due east of the stack showed barely
detectable
amounts
of plutonium.
An environmental
survey was begun
the following
day with a pickup
of vegetation,
soil, and water samples.
The soil analyses were not very definitive.
We acid-leached
them
and
separated
plutonium
by
our
then-routine
method
of bismuth
phosphate
- lanthanum
fluoride
co-
precipitation.
While we could detect
plutonium
by alpha
spectrometry
in some of the onsite
samples,
there were
other
alpha emitters
present,
the spectra
were smeared,
and we were unable to quantify
the results.
Of some
15 onsite
water
samples
collected,
pluto-
nium
was detected
in four
of them
at a maximum
of

GROSS ALPHA
ACTIVIT’Y RELEASED
100,8
,
1
,
s
s
‘
I
s
1
s
,
s
,
1

90 -

80 -

70 -

60 “
mCi

50 “

40 -

30 -

201

VIA POND 5

‘0:<,
,,,
,,,
,,,
,,,
n
“53
5455
56 57
58
59
60
61
62
63
64
65
66
67
68
6970
YEAR

Fig. IO

31

HIGH VOLUME. AIR SAMPLE 9-i I-57

00
,
,
u
1
1
I
I
I

4.76 MeV
90 “

80 “

70 -

60 -
5.15 MeV

50 -
m

40 -

30 “
4.18 MeV

20 “

10 -
,
[
o
10
20
30
40
50
60
70
80
90
100

CHANNEL

Fig. 11

0.5pCi/liter.
Of 35 offsite water samples collected
during
the
month
following
the fire, plutonium
contamination
was
noted
in
two
of
these
at a level too
low to be
statistically
valid--less than 0.1 pCi/Iiter.
Water
and
vegetation
samples
were
analyzed
by
extracting
with ether at that time. This method gave good
separation
of plutonium
and uranium
and a thin mount
for alpha spectrometry.
Figure
12 is a typical alpha spec-
trum
of a vegetation
sample. We detected
plutonium
on
most
of
the
vegetation
samples
collected
during
this
period
up
to
a maximum
of 600 pCi/kg
on 47 onsite
samples and 200 pCi/kg on 43 offsite samples.
Our
alpha
spectrometer
had
only
recently
been
acquired
and we had no pre-fire
data
on plutonium
on
vegetation.
Consequent Iy we were unable to estimate
how
much
of the observed
plutonium
was of fire origin. We
saw some plutonium
on samples taken from all directions
from
the
plant
but
the
maximum
were
to the
south,
downwind
at the time of the fire. The gross alpha activity
of these
samples
was somewhat
higher
than
our back-
ground
data
although
not
extremely
so. We could
not
detect
any
ground
contamination
on
the
plant
site by
direct
survey.
We concluded
from
these
measurements
that any offsite
contamination
resulting from the fire was
insignificant
and
there
were
no hot
spots
from
fallout
from the stack.
On May 11, 1969, a fire broke
out in building 776
and
eventuaUy
resulted
in multimillion
dollar
damage.

32

Although
there was some damage to one filter plenum the
building
essentially
maintained
its
integrity
and
little
plutonium
escaped. Contamination
was found on the roof
of
building
776
and
an adjacent
building
and
on the
ground on three sides of building 776. The roof contamin-
ation,
up to
10s CPM as measured
with
survey instru-
ments,
came from booster
1 exhaust.
Most of the ground
contamination
was caused by tracking during fire-fighting
operations.
Levels
up
to
10s CPM were
noted
on
the
ground.
Onsite air samples for the period May 9 through
12,
1969
ranged
from
0.03
to 0.31 pCi/m3
total
long-
Iived alpha. This is higher by an order of magnitude
than
we normally
observe
but
still well below the guide level
for insoluble
plutonium.
Offsite
air samples
showed
no
observable
elevation
of alpha activity.
This was also con-
firmed by the state of Colorado
Department
of Iiealthon
samples
taken
from
their
monitoring
net. The wind was
low and mostly from the northeast
during the fire.
Because
of re-entry
problems
we were
unable
to
ret rieve
our
exhaust
samples
until
May
15. The
three
samples
in
the
main
exhaust
showed
3.2,
21.6,
and
35.0 d/m/m3
total long-lived alpha for this period.
From
these data we calculated
a maximum
release via the main
exhaust
of
193 gCi during
the
144-h
period
of May 9
through
15. Booster
and dry air systems
samplers
shut
down
about
4 p.m. on the day of the fire due to power
loss.
Through
that
period
of
time
they
had
released
13 #Ci of Pu. Therefore,
release from the exhaust
system
was somewhat
in excess of 206 pCi (3.3 mg).

.

.
V O, 2E
IO-10-57

70
#
m
#
1
1
#
u
a
a
/ 5.15 Md/

60 -

50 -

40 -

COUNTS

30 -

20 -
4.18 MeV
4.76.MeV

10-

1

0
10
20
30
40
50
60
70
80
90
100

Liquid effluents
showed a maximum
level in pond 5
on
May
12,
1969
of
88 pCi/liter
gross
alpha
and
12 pCi/liter
in Walnut
Creek near Great
Western
Reser-
voir. During the month
of May, daily
samples
of Great
Western Reservoir showed a maximum
of 5 pCi/liter gross
alpha which is not elevated from normal readings.
Vegetation
samples
analyzed
radiochemicaUy
for
plutonium
ranged
up to 225 pCi/kg
of plutonium.
The
uranium
plus
plutonium
alpha
content
of these
same
samples
showed
no anomalies
from our routine
environ-
mental sampling program results in prior years.
As this information
was gradually
made public, the
Colorado
Committee
for
Environmental
Information
(Peter
Metzger,
Chairman)
and
a Rocky
Flats subcom-
mittee
of this group under Ed Marten took issue with our
conclusions
that no significant
amounts
of plutonium
had
been released during the fire. At a meeting at Rocky Flats
the subcommittee
argued
that
our air sampling
net was
not adequate
to detect
a channelized
release, that veget a-
tion was not a good sampling
medium,
that a land survey
for localized’’hot
spots”
should
be conducted,
that
our
water
data
showed
a
plutonium
buildup
in
Ralston
Reservoir,
and that
soil samples should be collected
and

CHANNEL

Fig 12

analyzed
for plutonium.
We took issue with some of their
points
but
did agree to conduct
a limited
soil sampling
program.
We collected
some 50 soil samples
in August,
1969 but
postponed
analyzing
them
or even developing
an analytical
method
for them
until
we had completed
our
other
environmental
samples.
In the
meantime
Ed
Marten and Stewart
Poet collected
soil and water samples
in the
area
and
analyzed
them
in their
laboratory
at
NCAR in Boulder.
Marten disclosed
his data in January,
1970 in a letter
to Glenn Seaborg.
Soil samples from
15
locations
mostly east of the plant ranged from 0.04 d/m/g
(his background
sample) to 13.5 d/m/g
of plutonium
and
seven water samples from 0.003 to 0.4 d/m/liter
of pluto-
nium.
Soil and water samples we had completed
by that
time were in general agreement
with his data.
The AEC sent Ed P. Hardy and Phil W. Krey from
NYO HASL to conduct
an independent
study of pluto-
nium contamination
in the area in February,
1970. Their
findings
are summarized
in HASL-235
(August,
1970),
They sampled
33 sites up to 40 miles distance
from the
plant
and
found
concentrations
ranging
up
to
2000 mCi/km2
(40 d/m/g)
offsite.
Using a 3-mCi/km2
contour
as
theti
lowest
readily
discernible
contour

33

(approximately
2 times background
from worldwide
fall-
out) they concluded
the contamination
from Rocky Flats
extended
east and southeast
up to 8 miles and contained
2.6 Ci (41.6 g) of plutonium
excluding
AEC-owned
land.
The
state
of Colorado
Department
of Health
also
conducted
a survey of plutonium
in surface soils offsite,
They
composite
25 surface
samples
from
each
of 13
segments.
SWRHL
analyzed
these samples
and found
a
maximum
of 24 d/m/g.
From
these
data
the Colorado
Department
of Health
estimated
0.3 Ci (4.8 g) of pluto-
nium as surface cent amination
offsite.
Using additional
data,
a group
of
Dow R and D
people estimated
offsite surface contamination
to be 7.6 g
of plutonium.
Although
Marten
had developed
his study because
he believed the May, 1969 fire had released large amounts
of plutonium,
it was soon apparent
that
the source
of
contamination
was not the fire but from a contaminated
area onsite, the 903 area previously
mentioned.
In the late 1950’s plutonium
processing
began gen-
erating
large quantities
of contaminated
cutting
oils and
solvents.
These
could
not
be shipped
as contaminated
waste nor processed
at the waste treatment
plant. While
technology
for handling
these wastes
and administrative
decisions
pursuant
were
being developed,
drums
of the
liquids
were stored
in a field beginning
in 1958.
Initial
plans
called
for
transporting
the
drums
to
the
waste
treatment
facility
for
processing
as soon
as necessary
equipment
was installed.
Rust-retardant
had been added
to the drums; however,
in 1964 it was determined
that it
would be necessary
to transfer the material to new drums
at
the
storage
site.
A small building
for
filtering
and
transferring
the liquids was erected
in 1966 and, in 1967,
the drum
removal
began.
The last plutonium-containing
drum was transferred
in January,
1968, and all drums had
been
removed
by June,
1968. Monitoring
of the storage
area
in
July
noted
levels
of
from
2 x 10s
to
3 x 107 d/m/g
alpha activity
and penetration
of the activ-
ity from
1 to 8 in. Fill was applied
the following
year to
help
contain
the
activity
and the actual
area on which
barrels
had been stored,
a 395 by 370 -ft rectangle,
was
covered
with
an
asphalt
pad
completed
in November,
1969. Additional
fill was added around
the pad in 1970
when soil samples ranging from tens to hundreds
of d/m/g
were obtained.
Soil stabilization
studies were started
to be
applied
to the entire area, and a revegetation
program was
begun.
From material
balance
calculations
it was estimated
that about 5000 gal containing
86 g of plutonium
(5.4 Ci)
had leaked.
We moved
one
of our
onsite
air samplers
to the
security
fence just
east
of the storage
area in 1963 to
monitor
the
area.
Figure
13 is a comparison
of the air
sample
data
from
this location
with the average of the
other
onsite
air samples from
1963 through
1970. These
data are total, long-lived alpha, not plutonium
concentra-

sample
station
is about
1/2 mi
from
the
nearest
plant
boundary
(which
is due
east).
Even
though
elevated
air
samples
were
observed
there
was
no
indication
of
the
extent
of offsite
contamination
occurring.
Referring
to
the arrows
on the figure,
from
left
to
right,
the
first
refers
to
the
point
in time
when
drums
were
first
observed
to be leaking, the second
to a period
of high winds
following
which
hot spots
were covered
with
dirt.
The
next
two demark
the time of the drum
removal operation.
The highest
point,
about
1/3 pCi/m3,
occurred
at the time vegetation
cover was removed
and
grading
started
preparatory
to pouring
the asphalt
pad.
The penultimate
point
at the right indicates
completion
of the asphalt
pad, and the final arrow indicates addition
of base course material around the pad.
In summary,
then,
plutonium
releases
to the en-
vironment
attributable
to
Rocky
Flats
can be broken
down as follows:

1. Cent rolled Releases
Airborne effluents
41 mCi=O.7g
Liquid Effluents
91mCi=l.5g

2. Uncontrolled
releases
1957 Fire
Maximum of 1 g
1969 Fire
0.21 mCi = 0.003 g
Wind-transferred
from
drum storage area
300-2600
mCi = 5-42 g

Obviously
the most dramatic
environmental
impact
has been from the contaminated
dirt transferred
by high
winds from the drum storage ~ea.
However, air-sampling
data
directly
downwind
indicates
that
applicable
guides
both
for occupational
exposures
onsite
and nonoccupa-
tional guides offsite
have not been exceeded
or approach-
ed.

.

L

tions. Even so, the average concentrations
arc well below
the
guide
for
insoluble
plutonium
of
1 pCi/m3.
This

34

I.0

TOTAL
LONG-LIVED
ALPHA
S-8
vs. Average
of Other
On-Site
Stations
I
I
i
1
I
I
I
I
I
I
I

I Readings from Station S-8

Y

0001 u
1
1
1
I
1
I
1
1
1
t
I
1
I

J:!
JUL JAN
JUL
JAN JUL
JAN
JUL
JAN
JUL
JAN
JUL
J/A# JUL
65
66
67
68
69

Fig. 13

ROCKY
FLATS
PLUTONIUM
RELEASES

I CONTROLLED

XI

.

AIRBORNE
EFFLUENTS
0.67g
42
mCi

LIQUID
EFFLUENTS
1.4
g
90
mCl

ACCIDENTAL

1957 FIRE
mg uptolg
-0.06CI

1969FIRE
3.2mg
-0.0002Ci

CONTAMINATEDSOIL

TRANSFERENCE
’42
g
-2.6
Ci

F@. 14
,.

35

DETERMINING
THE ACCUMULATED
DEPOSIT
OF RADIONUCLIDES

BY SOIL
SAMPLING
AND
ANALYSIS

by

E. P. Hardy and P. W. Krey

Health and Safety Laboratory

U.S.
Atomic
Energy Commission

New York,
N. Y.

ABSTRACT

Sdl
sampling and analysis is a feasible way to determine tha accumulated

amounts of
long-lived
radionuelides
that have deposited
on the ground.
The

Health and Safety
Laboratory
has measured ~r
and plutonium
isotopes in

soil samples to datermine
global and
regional deposition
patterns and inven-

tories. Site selection and representivity,
sampling, and analytical precision and

accuracy are dkussad
in this papar. It is shown that the precision of replicate

aliquoting
and analysis is the determining
factor in tha overall error associated

with soil semp/ing.

Introduction

Since
the
discovery
of plutonium
contamination
extending
outside
the Dow Chemical
Co. plant at Rocky
Flats,l
the;e has been a contagious
interest
in soil sampl-
ing. This has come about
primarily
because
of the failure
of
nuclear
plant
environmental
monitoring
systems
to
detect
chronic
low-level
releases
of radionuclides.
The
practice
of relating
total
alpha
or beta activity
measure-
ments
to the MlW’s has been satisfactory
from a regula-
tory
standpoint
but
it has not provided
the information
that
is now
demanded.
The questions
being asked today
relate to how much radioactivity
from a specific nuclide is
getting outside
the nuclear
plant boundary.
For the most
part,
satisfactory
answers
have not been given and plant
operators
have been
forced
to resort
to soil sampling
in
order to fmd out, as a first step, how much radioactivity
has accumulated
in the environment
from operations
to
date.
If adequate
air monitoring
and radiochemistry
were
carried
out
routinely,
soil sampling
should
play only
a
supplementary
role in a monitoring
program.
Soil is pri-
marily
useful
as an integrator
of initially
air-borne
long-
Iived radionuclides
that
have
deposited
on the ground.
Soil sarnpIing for this purpose
is not new to HASL since
we
have
used
this
method
periodically
since
1955
to

delineate
the
global
distribution
of fallout
%r
and to
inventory
the accumulated
deposit .23>4

Sampling and Preparation

It is easy but tedious
to sample soil, and analytical
methods
are straightforward.
The difficulty
is in selecting

a proper
site;
a site
which
represents
all of the radio-
nuclide
which
has fallen-out.
This does not
mean
a site
where, by some natural
process,
the radionuclide
is trans-
ported
horizontally
to another
spot once it is deposited.
In other
words,
we avoid those areas where accumulation
or depletion
can occur through
such phenomena
as flood-
ing or erosion.
These
kinda
of sites are easy to find. All
one has to do is sample along the base of a fence, under a
large tree,
in a drainage
ditch,
on the side of an ant hill
etc.
and
the results
might
depict
anything
except
what
actually
deposited
from
above.
There
are
conditions
where it may be easier to determine
what fell out over a
100 km2 area than over a one square kilometer
area. For
example,
we can do a global
inventory
for %lr
bysam-
pling less than one-tenth
of .a mz of ground
at only
100
sites
around
the
world.
It might
be considerably
more
difficult
to define local deposition
patterns
in a desert or
mountainous
area.

37

As for soil sampling in a locally contaminated
area,
HASL
demonstrated
that
the
Rocky
Flats
plutonium
could
be inventoried
by our methods.1
We were able to
describe
the
wntamination
pattern
as well and showed
that it extended
about
8 miles east and south
east of the
plant.
We see no reason why our soil sampling techniques
could
not
be applied
to any locally
contaminated
area
provided
that the contaminant,
was initially made airborne
in micron size particles
from the source.
Whenever
we
talk
about
deposition
of air-borne
debris the only meaningful
values are expressed
in units of
activity
or amount
per unit area. Soil sampling should be
carried
out
in such
a way
that
the actual
surface
area
sampled
is known.
‘Ilen
the entire
sample is weighed
in
the air-dried
state so that
the activity
per unit weight of
soil measured
can be converted
to area concentration.
Our sampling
and preparation
procedures
are well
documented
but
a brief
description
might
be helpful
here. We try to find flat grassed sites where we can take at
least ten, 3!4-in.-diam
cores in a stright line, spaced about
a foot
apart.
After
chying, the entire
sample
is crushed
and
blended.
Then,
about
3 kg are
passed
through
a
pulverizing
mill. This is the sample
from which aliquots
are taken for analysis.

Vertical
Distribution
of Radionuclides in Soil

Implicit
in the above discussion
is the need to take
the soil sample deep enough so that all of the radionuclide
deposited
is collected.
We know that in time any nuclide
initially
falling on the surface will migrate downward.
The
actual extent
of vertical penetration
will depend primarily
upon
the soil type,
but many other
factors
are involved
such as precipitation
amount,
chemical
form
of the nu-
clide, etc. At Rdcy
Flats we decided
to sam le down to
20 cm on the basis of our experience
with
k
r. It was
fortunate
that
we did because we found
that the Roc~
Flats
plutonium
was measurable
to
13 cm. Last fall we
took
depth
profde
samples
of
the
sandy
soil
at
Brookhaven
and
analyzed
them
for
137CSand
%r
in
addition
to %.
Table I expresses our results in terms of
the percentages
of the total
amount
deposited
for each
increment.
Cs-137
was measurable
down
to 21 cm and
%r
and ~
were detected
as far down as 25 cm. The
point
we want
to make here, however,
is that only 40 to
60% of the total
137CS,%r,
and ~
from nuclear tests
is in the
top
7 cm of soil. Similar
distribution
profiles
were
found
for
the
Rocky
Flats
plutonium.
1 If one is
interested
in measuring
all of the deposited
radionuclide,
we would advise sam@ng
from the surface to 30 cm.

Site Reproducibility

TABLE I

VERTICAL
DISTRIBUTION
OF FALLOUT
RADIONUCLIDES
IN BROOKHAVEN
SOIL
(FALL
1970)
‘

Depth
Percent of Total

(cm)

o-
7
7-11
11-15
15-21
21-25
25-30

59
42
57
26
30
27

11
15
11
4
8
3

0
2
1

0
3
1

sampling techniques
and did most of the %r
sampling for
HASL when he was with the Department
of Agriculture.
We have a considerable
body
of data
comparing
%r
in
soils at nearby
sites
2, 3, 4 as shown
in Table
II. The
average difference
between
pairs expressed as a percent
of
the
mean
was
calculated
for each
sampling
year.
The
deviations
range from 3 to 10%. We fmd these data very
useful
for convincing
skeptics
that
soil sampling
can be
used to determine
the cumulative
fallout for a rather large
area.
We now
have
some
information
on comparative
sites in the New York area for ~
deposition.
Table 111
shows
the
results
of 6 samples
taken
in three
different
locations.
The average _
deposit
is 2.3 mCi/krn2
with
a standard
deviation
of 13%. If it were not so difficult
to
fmd suitable
sites in the New York area we would prob-
ably have an even more precise value to report.

Analytical Preci3ion

Something
like a third of all soil samples analyzed
by HASL or contractors
are run as blind duplicates.
The
average percent
deviationa
between
aliquots
of prepared
samples
submitted
for
analysis
are shown
in Table
IV.
Again
we are expressing
the deviation
as the difference
between
pairs divided by the mean. For %Sr the errors are
less than
10% except
for the first year of sampling.
The
plutonium
analyses
were
done
in connection
with
the
Rocky
Flats study
and the average percent
deviation
was
20. We began plutonium
analyaes in soil only last year and
the procedure
requires
more skill at present
than does the
%r
method.
Under
these
considerations
this compara-
tively
larger
error
is understandable.
We are presently
analyzing
fallout
%%
in soil samples collected
through-
out the world and it appears
that the analytical
precision
will probably
average somewhere
between
10 and 20%.

b

The criteria for selecting
a site which represents
the

I

accumulated
deposit
in a particular
area have been
dis-
cussed
by
Alexander.
Dr.
Alexander
developed
the

38

TABLE
II

............

1*W

..---..--—.

19s9

~tlltc.
rl..

lU
1
~.
1 sit.

1964

..-..-...—-

1%6

61
71

G1
u

65

66

Is
u

30
27

1

4

.. ------

1%?

14
1s

1

A1..ka.
kctvu
5

-<th
Dakota.
llmtdbn
24

J.pn,
Takyo
s

m-it.
O.tnl
G

?.nun,
.2.2.
#

11.9.$.X.
$

ar.
stl.
ad-
G

U,d..
i..
Smu.blxy
2

Canada,
Akl.vlk
2Q

Cam.d.,
Ottam
9

-11<0,”1.,
m.
Anq.1..
>

mum
Africa.
Durban
2

u
z“l.nd.
IOS1U”,U9
,

mcny,
Vd.o
10

h-.
4Lr1
.
tit
.
.itm
Wwld

Ns.
.rrnt’,
s

12

lb

6

TABLE III

TOTAL 23%%IN SOIL AT NEW YORK
AREA SITES

TABLE IV

MEAN PERCENT
DEVIATION
BETWEEN
DUPLICATE
SOIL ALIQUOTS

Sampling
Period
Site

Dec. 1969
Fordham
Univ.
Jan. 1970
II
II
Jan. 1970
H
~J

July 1970
Bronx Botanical

Sept. 1970
Brookhaven
Sept. 1970
It

Depth
mCi per
(cm)
kmz

0-20
2.0
0-20
2.2
0-20
2.6

0-28
2.5

0-30
2.6
0-60
2.1

Avg. 2.3k 13%

Sampling
Year
Aliq. wt.
63)
No. of
pairs
Deviation
(%)
Isotope

19S6
,1957
1958
1959
1960
1963
1965-1967
1970
1970

%1
500
It

52
55
102
27
50
41
87
12
9

11
8
6
6
9
6
7
5

20

II

II
II

II

?1

II

II

II

II
.

1:0
100

39

Sample Size

When we first started
soil sampling about
15 years
ago, we took
20, 31A-in.-diam cores to
15 cm depth.
As
time passed and we had to go deeper
to get all the ‘%+,
we cut down
to 10 cores to minimize
the physical
exer-
tion of carrying
these large samples. The 1O+ore sample
represents
622 cmz of surface area. To test the reliability
of sampling
10 cores, Alexander
collected
duplicate
sam-
ples at about
10 sites throughout
the world. The average
deviation
turned
out to be 8%2 which convinced
us that
10 core samples
were adequate.
At Rocky
Flats we col-
lected duplicate
soils at 2 sites as shown in Table V. The
rocky
terrain
made
sampling
difficult
in some areas and
under
these non-ideal
conditions
we were satisfied
with
the agreement
between
duplicate
samplings.

Analytical
Accuracy

There
is no such animal as a primary
standard
soil
sample
for artificial
radioactivity.
In the first place,
no
two soil samples are alike and in the second place, there is
no
way
to add
a radionuclide
to
a sample
so that
it
represents
the chemical
and physical
form of the element
as it exists in the real world.
There
is such a thing as a
secondary
standard
soil sample. This could be represented
by a large quantity
of soil which has been dried, blended,
and pulverized
and aliquots
of which have been analyzed
on an inter-
and intra-laboratory
basis. We have such a
reference
soil which we are now using for our plutonium
fallout
study.
Table VI shows the available
results.
The
average value of 0.042 dpm/g is based on 13 results from
three
laboratories
using
100
and
100 g aliquots.
The
average deviation
is only 5% and we can see no sign~lcant
difference
among
laboratories
or aliquot
size. This will
become
a more
legitimate
standard
as time goes on and
other laboratories
report
their data. One day it may even
become a standard
in the true sense of the word.

TABLE V

DUPLICATE
SOIL
SAMPLING
IN THE

ROCKY FLATS AREA

Depth
mCl per kmz

Site
Sample
(cm)
23%
~r
—
.

6
1
0-20
2050
70
2
0-20
1500
65

7
1
0-20
490
69
2
0-20
440
64

TABLE VI

PLUTONIUM
IN REFERENCE
SOIL*

Lab

HASL
11
II
II

IPA
11
It
II
It

It

II

II

TLW
II

Aliquot

k)

1000
1000
1000
100

100
100
100

1000

dpm 23%
per g

0.043
0.042
0.042
0.041

0.049
0.042
().~~
0.044

100
0.041
100
0.041
100
0.042
000
0.042

000
0.041
000
0.042

Avg. 0.042 * 0.002
(5%)
——.
—________

*CoUected at Brookhavcn in October
1970 and consists of 100,
31%-in.diam cores to 2 in.

The Blank

A real blank
is a soil sample that is not contamin-
ated with the radionuclide
of interest.
We inherited
from
Dr. Alexander
a large quantity
of soil collected
in 1943. It
has served as a blank through
all of our %k programs arrd
now for our plutonium
work. The %
results are given
in Table VII. We conclude
from these data that contamin-
ation by laboratory
handling,
reagents, and other possible
sources under carefully controlled
conditions,
is not meas-
urable.

The Radiochemicd
Procedure
for Pu

Finally
we would
like to briefly
discuss the radio-
chernical
procedure.
It was developed
by Norton
Chu at
HASL to accommodate
100 g aliquots of the Rocky Flats
soils. It involves leaching
with
3 parts
nitric
acid and
1
part HCI.6 The plutonium
is separated
on an anion
ex-
change
column
and finally
electro-deposited
on a plati-
num
disc. The procedure
works for 1000 g aliquots
rdso,
with some minor modifications.
We have already
demon-
strated
that the acid leach quantitatively
removes
Rocky
Flats plutonium
from soil.1 Fallout
plutonium
can also be
acid leached
as we showed
in the same report.
We now

.

.

.

40

*

.

TABLE VII

PLUTONIUM
IN BLANK SOIL

Aliq. wt.
Lab
b)

HASL
100

IPA
100

100

100

100

100

1000

TLW
100
1000

‘vu
dpm per g

0.0003 + 100%

0.00008
f 100%
0.00008
* 100%

0.00005* 100%0
0.00003* 100%
0.00003* 100%
0.00002* 100%

0.0002 * 100%
0.00002* 100%

have
additional
supporting
data which is included
in the
summary
shown
in Table VIII. An analysis
of variance
indicates
that
the pairs of data are the same at the 9570
confidence
level. We felt a year ago that the question
of
whether
global fallout
plutonium
could
be acid leached
from soil was settled.
This is simply a reiteration
of our
position
with some additional
evidence.

Fallout Pu-239

Dr.
Harley
has
referred
to
our
present
study
to
inventory
the global
deposit
of SNAP-9A
-.
One side
benefit
of this
work
will
be a general
picture
as to how

-u
is distributed.
The
figure
shows the accumulated
deposit
of ‘%%
at sites sampled
in the
United
States

Sampling
year

1969
1958
1967
1970
1970
1969
1969
1970
1970

during the falI of 1970. There are no surprises or obvious
anomalies.
These
are about
the levels one would expect
from
the
weapons
tests
conducted
so far.
The heavier
precipitation
areas and the mid-latitudes
rhow the higher
deposits just as we find with ‘%.
We know there are local
areas
of
contamination
such
as at Roclg
Flats.
On a
country-wide
scale,
however,
if there
is any plutonium
that has been or is being released from a nuclear facility,
it has not perturbed
the accumulated
deposits
from test-
ing enough to detect it.

Conclusion

We have discussed soil sampling
for the purpose
of
determining
the accumulated
deposit of initially air-borne
radionuclides
such as %lr
and plutonium.
Site represen-
tivity,
depth,
sample
size, and
analytical
prec~lon
and
accuracy have been considered.
We are convinced
on the basis of our quality
control
experience
to date that
the precision
of replicate
aliquot-
ing and analysis is the determining
factor
in the overall
error associated
with soil sampling.

References

1. P. W. Krey and E. P. Hardy, “Plutonium
in Soil Around the
Rocky Flats Plant;’ HASL235,
August 1, 1970.

2. L. T. Alexander, et al., “Strontium-90
on the Earth’s SUrface;’
TfD-6507, February 1961.

3. E. P. HardY, et al., “Strontium-90
on the Earth’s Surface II,”
TID-17090, November 1962.

4. M. W. Meyer, et al., “Strontium-90
on the Earth’s Surface IV:’
TID-24341, May 1968.

5. J. H. Harley, editor, “Manual of Standard Procedures, D-W,”

NY04700.

6. N. Y. Chu, “Plutonium
Determination
in Soil by Leaching and
Ion-Exchange
Separation;’
Anal.
Chem.,
43,
No.
3, pp.
4494S2,
March 1971.

TABLE VIII

ACID LEACH Vs. COMPLETE DISSOLUTION
OF ‘%
IN SOIL

Depth
dpm ‘%
per g

Site
(cm)
Ieach
comp. sol’ n

New York
5-20
0.0044

Illinois
0.0042

0-15
0.0051
0.0047

New York
0-20
0.017
0.017

Brookhaven
o-5
0.042
0.042

Rocky Flats
0-20
0.060
0.080
New York
o-5
0.094
0.091
New York
o-2%
0.21
0.24

Rocky
Flats
0-20
3.08
3.18

Rocky Flats
0-20
17.5
16.0

41

\

<
/
\..
1
“\.
t.,
i
I
\

i.._-"-
\.
i
4

1“
--4=;-
---

\
---
--.
-’--.--q
.

“\

.0.7

I
I
I
‘\.

“-u

#..h” e a.b

0.9

.

42

Fig. 1

Accumukted
deposit of 239A at sites sampled in the United States dun”ng 1970.

ANALYTICAL
TECHNIQUES
FOR
THE
DETERMINATION

OF PLUTONIUM
IN ENVIRONMENTAL
SAMPLES

by

N. A. Talvitie

Western Environmental
Research Laboratory

Environmental
Protection Agency

Las Vegas, Nevada

ABSTRACT

Techniques used by the Western Environmental
Research Laboratory
for

improving the accuracy and economy
of plutonium
determination
in environ-

mental samples are presented. Ignited soil, air filter, and vegetation samples are

prepared for
analysis by
rapid, total dissolution methods in disposable poly-

propylene
beakers. Plutonium
in sea water is concentrated by coprecipitation

on ferric hydroxide.
High adsorption
efficiency and separation from
thorium

are obtained
by
ion exchange separation of plutonium
as the chlorocomplex

ion,
Hydrogen
peroxide
is used both
for
stabilization
of plutonium
in tie

quadrivalent
state during
adsorption
and for
reduction
to the trivalent state

during
elution.
Sample
sources for
alpha
spectrometry
are
prepared
by

60-minute
electrodepositions
from
ammonium
tilfate
media
on
electro-

polished stainless steel planchets mounted
in low-cost, disposable cells. Count-
ing data are converted
by
a computer
program
to
a report
format
giving

activity of
‘?%
and
-u
per sample unit, deposition
per square kilometer,

and error terms. The mean overall yield from
environmental
samples is 94Y0.

The
full
width
at
half
maximum
resolution
is 37.5 keV at
12.5 keV
per

channel
and
21%
counting
efficiency.
The
minimum
detectable activity
is

10 f Ci of ‘*u
for a 1000-minute count.

--------
.— --- .. .... .. .. .. ... ... ... . .. ........

Introduction

The Technical
Services Program
of the Western En-
vironmental
Research
Laboratory
has analyzed
environ-
mental and biological
samples for a number of plutonium
studies.
Among these were analyses of air, water, and soil
samples
as assistance
to the State of Colorado
in studies
of the Rocky
Flats area; of air and soil samples collected
at Bikini Island; and of air, water, soil, precipitation,
and
vegetation
samples collected
in the offsite areas surround-
ing the
Nevada
Test
Site. Although
the analysis
of sea
water
is primarily
a readiness
program
for incidents
in-
volving plutonium-containing
devices, the laboratory
has
provided analyses following one such incident.
The analytical
process applied to samples consists of
the following
operations,
which
can be performed
inde-
pendently:
sample
control,
preanalysis
preparation,

dissolution
and
concentration
operations,
ion exchange
separation,
electrodeposition,
alpha
spectrometry,
and
computer
computation
of results.
Aside from the consid-
erations
of
acculacy
and
economy,
the
selection
and
development
of techniques
has been
to provide
a single
process for all types of environmental
and biological sam-
ples.
Some
of the
techniques
have
been
reported
pre-
viouslyl’2
and are presented
below
as summaries.
Tech-
niques that differ from these are presented
in detail.

Sample Control

The sample
is assigned a serial number
and all per-

tinent
information
is
coded
on
an
IBM
card
which
accompanies
the sample
to the appropriate
laboratory.

43

Pre43mlysis Preparation

At
present,
two
types
of environmental
samples
receive processing
independent
of the plutonium
labora-
tory.
Vegetation
samples
are ignited
in large muffle
fur-
naces and the pulverized ash is submitted
for analysis. Soil
samples are air-dried and sieved with a 10-mesh sieve. Any
friable material
and loose aggregates of soil in the oversize
fraction
are crushed
in a mortar
and passed through
the
sieve. Sticks and gravel retained
by the sieve are discarded.
A 30- to 40-MI aliquot
of the 10-mesh fraction,
obtained
by repeated
mixing
and splitting
in a riffle, is submitted
for analysis
in a four-ounce,
wide-mouth
jar. When soil
and
sediment
samples
have
an unusually
high
organic
content,
it
is more
convenient
to
dry
and
ignite
the
sample before an aliquot is taken. Preanalysis
preparations
performed
in the analytical
laboratory
are discussed
in
conjunction
with
sample
dissolution
and
concentration
techniques.

Dissolution
and Concentration
Operations

AU samples are prepared
for the ion exchange separ-
ation of plutonium
by methods
which, in effect,
provide
for total
dissolution
of the
sample.
The preparation
is
simplified
by the use of the 6&l azeotropic
concentration
of hydrochloric
acid as the final solvent.

Soil. The aliquot
of 10-mesh soil is dried in an oven
overnight
at
110°C
and ground
to
a fine powder
in a
centrifugal
ball mill. It is then returned
to the sample jar
and mixed by rotating
the jar mechanically
end-over+nd.
A one-gram aliquot is ignited in a porcelain crucible,
transferred
to a 100-rnl disposable
polypropylene
beaker,
and spiked
with
‘Pu.
A mixture
of hydrofluoric
and
nitric acids is added and evaporated
to dryness on a steam
bath
with
a specially-built
top
of l/4-in.
polyethylene.
The top has 24 holes which allow the beakers
to sit with
two-thirds
of their depth into the bath. The evaporation
is
repeated
with a smaller volume of hydrofluoric
and nitric
acids to ensure
complete
decomposition
of the soil and
volatilization
of fluosilicic
acid. Nitrate
and fluoride
are
removed
from
the residue
by evaporating
successive vol-
umes of hydrochloric
acid to dryness in tie
beaker.
The
residue is then dissolved in m
hydrochloric
acid contain-
ing a few
drops
of
hydrogen
peroxide.
The
peroxide
reduces any hexavalent
plutonium
which might have been
produced
during the decomposition
process.

Vegetation
Ash and Air Filters.
Vegetation
ash and
air fiiters are decomposed,
spiked with ‘Pu
and prepared
as 6~
hydrochloric
acid solutions
in the same manner
as
soil. One g of vegetation
ash is weighed
directly
into
a
tared
polypropylene
beaker.
Glass-fiber
falters are folded
into a wad, ignited in cupped stainless steel planchets,
and
transferred
to
the
beaker.
If
the
weight
of the
fiiter
exceeds
1 g, a section
is cut from the falter for analysis.

Composites
are made by cutting
circles from each filter to
represent
a known
fraction
of the filtering
area. Filters
composed
of organic
materials
are ignited
in platinum
beginning
with
a cold
furnace.
Any
amount
of filter
material can be handled
provided that the total ash weight
does not exceed
1 g.

Water and Precipitation.
Sea water and saline water
samples
are
acidified
with
hydrochloric
acid.
Plutonium-236,
iron carrier,
and hydrogen
peroxide
are
added
and
the sample
is heated
to decompose
the per-
oxide.
The iron acts as a cataIyst
to decompose
organic
matter
while the valence
equilibrium
in the presence
of
hydrogen
peroxide
serves
to
interchange
the
internal
standard
with
the environmental
plutonium.
The pluto-
nium is carried
on a ferric hydroxide
precipitate
which is
redissolved
to give a @l hydrochloric
acid solution.
Any
insoluble
residue
is separated
and then solubilized
by an
abbreviated
version of the soil method.
Plutonium
in fresh water and precipitation
samples
can
also
be
concentrated
by coprecipitation
on
ferric
hydroxide.
This
is convenient
if the
sample
has been
faltered
to
determine
the
soluble
plutonium
separately
from that associated
with the suspended
solids. For total
plutonium
in unfiltered
samples, the sample is evaporated
to dryness,
wet-ashed
with nitric acid and hydrogen
per-
oxide,
and that portion
of the residue which is insoluble
in 6&f hydrochloric
acid is solubilized
by the abbreviated
soil method.

Ion Exchange Separations

Apparatus.
The ion exchange equipment
consists of
a bank
of 24 columns
in a hood
specially-designed
to
carry
off
the
acid
fumes.
The
units
are commercially
available and consist of 14.5 -mm-i.d. tubes having integral
reservoirs
and stopcocks
with Teflon
plugs. The catalog
item is modified
by adding
a sealed-in,
coarse glass frit.
The
columns
contain
a 20-ml
volume
of anionic
resin
capped
with
a layer of fine silica sand. The sand enables
reagents
to be added without
particular
care and, because
the capillarity
of the sand acts as valve to stop the flow,
the operator
is free to spend
time on other
phases of the
analysis.

tilumn
Operation.
The 6?4 sample solution
is ad-
justed
to 9~
by adding an equal volume of concentrated
hydrochloric
acid.
Hydrogen
peroxide
is added
to shift
the equilibrium
in favor
of the quadrivalent
state.
The
solution
is faltered into
the reservoir
through
a plug of
glass wool in the stem of a disposable
funnel.
The filtra-
tion
removes
barium
chloride
which
precipitates
from
glass fiber filter samples and sodium chloride which occa-
sionally precipitates
from evaporated
water samples. After
passage
of
the
sample
solution
through
the
resin,
co-
adsorbed
iron
and
uranium
are selectively
eluted
with
nitric
acid. The nitric acid eluate can be reserved for the
determination
of ‘sFe and uranium.

G

44

The plutonium
is eluted
with a 1.2~
hydrochloric
acid-O.6%
hydrogen
peroxide
reagent,
The peroxide
in
dilute acid shifts the equilibrium
in favor of the trivalent
state
and has
another
advantage
in that
no nonvolatile
impurities
are introduced.
A O.S-ml volume
of concen-
trated
sulfuric
acid is added
to the eluate,
which is then
evaporated
overnight
on a low-temperature
hot plate. No
fuming of the sulfuric acid or wet-ashing is required.

Electrodeposition

Apparatus.
The
disposable
electrodeposition
cells
are constructed
from
linear-polyethylene
liquid scintilla-
tion vials and hold a 3/4-in.
stainless steel planchet.
The
cell supports
and cathode
contacts
are 1/8-in. potentiom-
eter shaft
locks attached
to machined
Lucite bases with
non-insulating
banana-plug
jacks.
Twelve
electro-
deposition
units
are operated
in parallel
from
a single
power
supply.
A storage
battery
automatically
supplies
current in case of a power failure.

Electropolishing.
The planchets
are polished
to a
mirror
finish while mounted
in the cells using a reversed
current
of
1.2 A for six minutes.
The
electropolishing
electrolyte
is an adaptation
of a formula
containing
phos-
phoric
and
sulfuric
acids
which is used industrially
for
polishiiig stainless steel.

Electrodeposition.
The
sulfuric
acid
solution
of
plutonium
is
diluted
and
neutralized
to
give
a
1~
ammonium-sulfate
electrolyte
having a pH of 2.0 to 2.3.
The deposition
is essentially
quantitative
in 60 minutes
of
electrolysis
at 1.2 A.

Alpha Spectrometry

Apparatus.
The
counting
system
has eight
silicon
surface-barrier
detectors.
Two detect ors are mounted
in
each of four vacuum chambers.
The bias voltages for each
pair
of detectors
are provided
by dual power
supplies.
Each detector
of the pairs has its own preamplifier,
linear
amplifier,
and biased amplifier
but the pairs of signals are
brought
into
dual input 400-channel
analyzers
operating
in the multiplex
mode. The data from the four analyzers
feed into a single digital printer through
solenoid-perated
banks of switches.

Spectrometry.
The ener~
range is 3.5 to 6.0 MeV
in 200 channels
which covers most of the alpha emitters
of interest.
The
plutonium
peaks
appear
in the second
100 channels of the 200-channel
spectrum.
The resolution
is three channels at 12.5 keV/ch or 37.5 keV fuU width at
half maximum
and the mean counting
efficiency
is 22%.
The counts
in 16 channels
are summed
for each of the
plutonium
isotopes.
Low-level
samples
are
counted
overnight
for
1000 min and higher level samples during the day for 400

min. When the sample load is light, low-level samples are
counted
for 1400 min. The detection
limit for’%
with
1000-
to
1400-min.
counts
is
10 fCi at
two
standard
deviations.
The detection
limit of au
is 20 fCi because
of the higher background.

Computer
Computation

The
sample
data,
sample
and
blank
counts,
and
calibration
data
are coded
for the computer.
The com-

L
uter is programmed
to give a printed
report of’%
and
activity
per
sample
unit,
deposition
in
soil per
square kilometer,
two sigma error terms, and the percent-

‘Pu
The
yield
serves
as a quality
control
age yield
of
.
over
the
sample
preparation,
ion
exchange,
and
electro-
deposition
techniques.
The yields
are generally
over 90%
and average 94%.

Conclusions

Techniques
have been selected
to improve
the ac-
curacy
and economy
of the analytical
process.
Total dis-
solution
methods
insure
that
all
of
the
plutonium
is
exchangeable.
Low-cost,
disposable
equipment
minimizes
cross-contamination
and eliminates
the need for involved
decontamination
procedures.
Electropolishing
of the dis-
posable
stainless
steel planchets
results in a scrupulously
clean
and bright
surface
at low cost.
The ion exchange
and electrodeposition
methods
give high chemical
yields
and essentially
weightless
sample
sources
which contrib-
ute to counting
precision,
The
decomposition-dissolution
procedure
for soil,
air falters, and vegetation
ash requires
less than two man-
minutes
of attention
per sample and can be scaled up to
handle
2.5 or 4 g of sample
at little
additional
cost by
using
correspondingly
larger
disposable
beakers.
When
samples larger than these are required
in order to integrate
a non-uniform
distribution
of plutonium,
an aliquot
can
be taken
for additional
processing
after
the sample
has
been decomposed
sufficiently
to interchange
the internal
standard
with
the
enviornmental
plutonium.
Sea water
samples up to 10 liters in volume can be analyzed
without
modification
of the basic procedures.
Reference
1 con-
tains a procedure
for the analysis of 10-g samples of coral
limestone
soil. The sensitivity
is adequate
to detect back-
ground
levels due to worldwide
contamination
from nu-
clear
testing
in
1 g of
surface
soil or in an air falter
representing
500 cm3
of air. One-liter
samples
are ade-
quate
for
the
determination
of
plutonium
in potable
water.
Because the operations,
other
than sample prepara-
tion,
are
identical
for
all types
of environmental
and
biological
samples,
technicians
can conduct
all phases of
the analytical
process
after
a short
training
period;
and,
because
the operations
can be conducted
independently,
peak
sample
loads can be handled
by temporary
assign-
ment
of personnel
but
do not
require
a corresponding
increase in space and equipment.

45

References

1. N. A. Talvitie, “Radiochemical Determination
of piutonium in
2. N. A. Talvitie, “Electrodeposition
of Actinides for Alpha Spec-
Environmental
and Bilogical Samples by Ion Exchange.” Pre-
trometry.”
Presented at the 24th Annual Northwest
Regional
sented
at
the
American
Industrial
Hygiene
Conference,
Meeting of the American Chemieal Society, Salt Lake City,
Toronto,
Ontario,
Canada May 24-28.
1971. Submitted
for
Utah
June
12-13.
1969.
Submitted
for
publication
in
publication in Anulyticuf
Chewr&ry.
Analytical
Chemis&.

46

.

.

SAMPLING
AND
ANALYSIS
OF SOILS
FOR
PLUTONIUM

F. E. Butler, R. Liebermen, A. B. Strong, and U. R. Moss

Eastern Environmental
Radiation
Laboratory

Environmental
Protection Agency
Montgomery,
Ala.

ABSTRACT

This Pe%r
describes the progress in analysis of
soils artificially spiked
with plutonium,
soils containing particulate plutonium
deposited from a proc-

essing plant,
and
soils containing
fallout
plutonium.
The
emphasis is on
distribution
of
the
actinide determined
after both fusion
and acid leaching

techniques.

The residue from
multiple evaporations of soil with hydrofluoric
acid is
fused with
potassium f Iuoride and potassium pyrosulfate,
dissolved in dilute

sulfuric acid, and the solution
evaporated to remove fluorides.
Plutonium
is

then extracted with a hydrochloric
acid solution with tri-isootylamine
(TIOA)

and
stripped from
TIOA
with dilute
acid. Plutonium
is coprecipitated
with

LaF~, the precipitate filtered onto a 0.2-# @ycerbonate
filter membrane, and

the plutonium
counted in an alpha spectrometer.

Recovery, indicated by 23GPutracer added to each sample, is 75 + 6% for
5-g soils. Recovery is higher for smaller samples. Assays of five interlaboratory

cross-check soils in the range 0.5 to 16.0 pCi/g yielded an average error of only

3.6% by this method.

..- .......- —-.. --_-.
. . ... ............. .... .. ....-

Introduction

There are a number of problems
associated with the
analysis
of
plutonium
isotopes
in
soil
samples.
These
problems
can be attributed
to one or both of the follow-
ing conditions:

.
The plutonium
may be of a refractory
nature and
not easily separated
from the soil matrix.

.
The mode of distribution
of the plutonium
could
have produced
erratic
and nonuniform
dispersion
of the
radionuclide
in the soil.

A number
of fusion
procedures
have been devel-
oped to insure dissolution
of refractory
components,
in-
cluding
plutonium,
from
soil samples.
These
methods,
however, are limited to soil sample sizes of 10 g or less.
Analysis of small soil samples by fusion can result in
misleading
data
dependent
upon
the
degree
of

nonuniformity
of the plutonium
at the sampling site. To
overcome
this
difficulty,
larger
soil samples
have been
leached with various acid mixtures.
This paper describes
a fusion procedure
used at this
Laboratory
for plutonium
analysis in soil samples. Results
of the
procedure
are compared
with
various
acid leach
procedures
performed
on identical
soil samples.

Experimental

Fusion
of Soil.
Initial
experiments
using the rea-
gents
potassium
carbonate,
sodium
carbonate,
sodium
tetraborate
decahydrate,
barium
sulfate,
potassium
hy-
droxide
and
others
in various
combinations
were
not
successful
in this Laboratory.
The reagents
showing
the
most
promise were those “used by Silll
for “@b
analysis
of soil. Variations
of these reagents yielded a fused sample
that was completely
soluble in 6~ HC1. The procedure
is
as follows:

47

I.
Add 5gof
dried,
sieved, and muffled
(550°C)
soil toa
teflon beaker. Add2MPu tracer.

2. Add 35 ml of 28&l I-IF and evaporate
to dryness
at low heat.
Repeat
three
more
times to volatilize
the
silica. Finally, add 15 ml of 12FJ HC1and evapxate.

3. Transfer
the
powdery
residue
to a 50-rnl plat-
inum crucible with the aid of a pliceman.

4. Add
4 g of KF.
Place a platinum
top
on the
crucible
and fuse over a reeker
burner
for 30 min. Add
7.5 g of Kz Sz 07 and fuse for an additional
30 min.

5. Cool the crucible
in an ice bath,
add
15 ml of
12FJ HC1 and
evaporate.
Add 30 ml of water,
heat
and
transfer
to a beaker.

6. Rinse the crucible
with
a portion
of 200 ml of
6~ Hz S04 added to the beaker. Evaporate
past the white
S03- fumes to remove all traces of F-.

purification
with
TIOA.
The liquid
ion exchanger
tri-isooctylamine
(TIOA)
re~rted
previously
was used
to
separate
the
plutonium
isotopes
from
calcium
and
other
trace elements
in soil as well as natural
uranium.
The procedure
is as follows:

1. After
removal
of fluorides,
dissolve
the residue
in 6~ HC1 with
heat,
Use the
total
volume
of 400 ml
6~ HC1, including
rinse,
to
transfer
the
solution
to a
separator
funnel.
Add 10 drops of SO%H202
to adjust
the Pu to valence (IV).

2. Add 25 ml of 10% TIOA-xylene
and shake brief-
ly. Invert
the funnel
and release the pressure.
Shake the
solutions for one min.

1. Dissolve
the
wet-ashed
residue
in
10 ml
of
1~ HC1, heating to about 60”C.

2. Cool the solution
to room temperature
and add
1 drop of 50% Hz 02 to adjust Pu to valence (IV).

3. Add
0.1 mg of
lanthanum
(lanthanum
nitrate
dissolved
in 1~ HC1) and 2 ml of 3~ HF and allow the
precipitate
to form for 30 min.

4. Filter
in a Millipre
apparatus
onto
a 25-mm
0.2 B membrane.
Wash the beaker
with water then with
alm”hol.

5. Mount
the
hesive tape attached

6. Count
the
spectrometer.

filter membrane
on double-faced
ad-
to a 30-mm planchet.

sample
for
1000 min
in the
alpha

7. Calculate
the
quantity
of
plutonium
isotopes
and correct
for the recovery
of the known
2%J
added
initially.

Leaching
Experiments.
A soil sample
was spiked
with: 23%.
The sample was dried, muffled,
and thorough-
ly mixed and analyzed by the fusion procedure.
Duplicate
leaching
experiments
were
conducted
with
six solutions.
One-g samples
of soil were heated
to
boiling
with
10-ml volumes
of leach solution
and then
allowed
to digest for one h. They were then filtered
and
the filters washed
with
hot water until the total volume
for each sample was 20 ml. One-ml aliquots were analyzed
by
liquid
scintillation
counting.
Results
are shown
in
Table I. Note
that the HC1 leaches were more complete.
Subsequent
tests
on a variety
of soik,
including
those
mentioned
in the next
section,
showed
that HF is often
required for complete
leaching.
3. Drain
and
discard
the aqueous
solution.
Rinse
the organic
layer with
25 ml of 61J HC1 and discard
the
rinse solution.
TABLE I
4. Strip the Pu from TIOA with two 25-ml volumes
of
4N HCI-O.051J HF,
shaking
for
two
min
each
strip.
(Urafium
may
then
be stripped
from
the
TIOA
with
O.1~ HC1and analyzed separately.)

5. Add 10 ml of 16~ HN03
to the combined
strip
solutions
and evaporate
to dryness.
Further
wet ash the
residue with 5 ml of 12~ HC1 plus 5 ml of HC104.

Coprecipitation
and Counting.
Plutonium
is copre-
cipitated
with a trace amount
of LsF33 and filtered onto
either
a polycarbonate
fiiter membrane
(IWclepore)
or a
solvinert
membrane
(MWipore).
The
automatic
low-
background
alpha
spectrometer
was described
previous-
ly.4 The procedure
is as follows:

LEACHING EXPERIMENTS
OF SOIL CONTAINING
1700 DPM 239PU PER GRAM

Leach
Solution

Water
41J HC1
12~
HC1
IIJ HF
281J HF
41j HCL
-lIJ
HF

239Pu dpm/gram
Sample 1
Sample 2

0
0
1520
1600
1520
1600
0
0
80
220
740
720

.

A.

.

48

.

..

Results and Discussion

Figure
1 shows the alpha spectrogram
obtained
by
analysis
of a soil through
the fusion procedure
using the
polycarbonate
membrane.
Note
the
good
resolution
of
‘Pu,
‘%,
and
‘%,
which
allows
the
quantitative
determination
of the isotopes.
Table 11shows good precision and accuracy of aml-
ysis of five interlaboratory
soils by the fusion
method.
Although
the
fusion
method
and
subsequent
chemical
separation
is described for 5-g samples of soils, it has been
employed
for different
quantities
of soil. The fusion of
more than
10 g of soil appears impractical
with this pro-
cedure.
Analysis
of
20
enviornmental
soils
from
Mont-
gomery,
Alabama and Cape Kennedy,
Florida, resulted in
“%
recovery
of
75 * 6%. These
5-g samples
assayed
between
less
than
sensitivity
(.o3
dpm)
to
0.08 dpm/g ‘S.
One concern
in analysis of soil is the distribution
of
plutonium
particles
and, therefore,
the proper techniques
for sampling and the optimum
amount
of sample required
for representative
analysis. To investigate
these factors,
a

236PU
\

238,

239PU

\

Al

I
50

soil was obtained
from
a nuclear
processing
plant where
‘%
had been deposited
in particulate
form by accident
approximately
one
year
before
receiving
the
soil. The
particles
had been
covered
with approximately
12 in. of
fresh soil during the year prior to sampling.
The soil was dried, muffled
at 5500C and thorough-
ly mixed prior to analysis of 21 l-g samples by the fusion
method.
The
recovery
of
2%%
was 81.6 + 8.3% with
maximum
and minimum
recoveries
of
!)9Y0
and 64%. The
23%
in
the
soil was 0.57 * .40 dpm/g;
however,
with
maximum
and minimum
assays of 1.72 and 0.25 dpm/g.
The relative
standard
deviation
was * 70% compared
to
only 8% for the added tracer.
Twenty-g
batches
of the
above
soil were leached
with a total volume of 200 ml of solution
in the manner
described
in the Experimental
Section.
Ten-ml
aliquots,
representing
1 g of
soil,
were
analyzed
by
the
TIOA
exchange
procedure.
Results are shown in Table III. Note
that
these
analyses
show the HC1-HF leaches yield 23%
assays very close to the mean of the 21 fusion assays.

summary

1. A fusion method
is described
which yields accur-
ate plutonium
results for small samples (1 to 10 g) of soil.

2. The
distribution
of
particulate
plutonium
de-
posited
accidentally
on soil can vary almost tenfold
from
gram to gram.

3. Analysis of a relatively
large portion
of the par-
ticulate
soil after acid leaching results in less variation
in
replicate
analysis
than
the
analysis
by
fusion
of
1-g
aliquots.

4. No leach experiments
were performed
on actual
atomic
debris plutonium;
therefore,
no claim is made that
the highly refractory
plutonium
in fallout is soluble in the
various leach solutions.

Acknowledgment

The authors
wish to express
their appreciation
for
the
technical
assistance
of
Miss J.
Favor,
Mrs. E. W.
Pepper,
and Mrs. M. W. Williams in performing
the labora-
tory experiments.

+

100

Channel
Number

F&. I.

49

TABLE II

RESULTS OF PLUTONIUM
IN SOIL CROSS CHECKS

.

Sample
Number

1

EERL (pCi/g)
KNOWN (pCi/g)
ERROR,
ms~
23%
(%)

238pu
239pu
Z3.sh
*+U

.407
15.80
.319
15.90
.306
15.60

Avg.

2

Avg.

3

.344
15.77
.26
15.68
24.5
0.6

-.
.031
.032
.030

-.
.031
.031
.-
0.0
---

—
2.43
2.56
2.31

Avg.

4

.-
2.43
2.24
8.5
...
.-

...
16.98
16.36
15.69

Avg.

5

.-
16.34
15.59
4.8
...
..-

...
0.52
0.52
0.45

Avg.
...
0.49
0.47
4.2
...
-.

Avg.
3.6

TABLE 111

LEACHING
TESTS
USING
SOIL
CONTAINING
PARTICULATE
239pu

239Pu Assay
(dpm/g)
!%mule 1
sample
2

0.22
0.22
0.57
0.42
0.64
0.69

3.
R. Lieberman and A. A. Moghissi, Health Phya. 15, 3S9
(1968).

4.
H. L. Kelley, R. E. Shuping, R. H. Schneider, and A. A.
Moghissi,Nuclear Instruments and Methods 70, 119 (1969).

Leach
Solution

4~
HC1
4~
HCI - 1~ HF
4~
HC1 - 2~
HF

References

1.
W. Sill and C. P. Wiilia,AnaL Chem. 37 No. 13166 (1965).

2.
F. E. Butle:, Health Phys. 15, 19 (1968).

50

USE OF PLUTONIUM-236
TRACER
AND
PROPAGATION
OF
ERROR

w

Claude W. Sill

Haalth Sarvices Laboratory

U. S. Atomic
Energy Commission

Idaho Falls, Idaho

ABSTRACT

The use of ‘Pu
tracer to make yield corrections in the determination
of

both
‘8Pu
and
‘9Pu
is discussed, both
from
the theoretical and practical
points of view.

The consequence of using too-small quantities of ‘Pu
tracer is that the

uncertainty
in the yield determination
beconws much greater than the uncer-

tainty
in the total count
of
plutonium
in the sample. If large quantities of
ZsGputra=r
are used to improve the statisticsof the yield determination,
other

problems are introduced; these are discussed.

Plutonium-236
tracer
has been used almost univer-
sally for several years to make
yield corrections
in the
determination
of both
‘%s
and ‘%.
Although
it is of
great
assistance
when
used properly,
many
investigators
have
apparently
considered
the
ability
to
correct
for
chemical
inadequacies
to be an adequate
substitute
for
good chemistry,
even when the yield goes as low as 10%.
There are several problems associated
with its use, none of
which have even been mentioned
in any of the articles on
the determination
of piut onium so far examined.
As should be well known,
the statistical
uncertainty
in the determination
of the yield must be passed on to the
determination
of the nuclide
being sought in the sample.
Yet,
few
analysts
seem
to
consider,
at
least
in their
published
works, the effect
of quantity
of tracer used on
the
sensitivity
and
accuracy
of
the
determination.
A
widely used method
of error propagation
shows that the
fractional
error in the value of the nuclide being sought is
equal to the square root of the sum of the squares of the
fractional
errors
in each
of
the
independent
variables
involved.
If X, Y, and Z are the total
counts
obtained
in
the energy intervals
for the ‘%
being sought,
the ‘f%
recovered
through
the
procedure,
and
the
-
in the
standard
from the same quantity
of tracer, respectively;
g,
Ez, and t are grams of sample, counting
efficiency
used in
the
standardization,
and
time
in minutes,
respectively;
and BX, BY and BZ are the respctive
background
counts

G

or other corrections
for the same counting
time, then

(x - B,)
(z -%)
23@udpm/g=V_
BY) “gEzt
“

In other words,
the concentration
of 23%s in the sample
is simply
the
ratio
of net
counts
of 23~
to _
re-
covered
multiplied
by the dpm/g of *“Pu added as tracer.
It should
be noted
specifically
that
once the concentra-
tion of 2%Pu used has been determined,
neither
counting
time,
counting
efficiency,
nor
errors
therein
have any
effect
on
the
accuracy
of
the
determination
except
as
they affect
the statistical
errors resulting from total num-
ber
of
counts
obtained.
Elimination
of
the
effect
of
changes in counting
efficiency
is particularly
important
in
routine
work because
a significant
source of inaccuracy
is
the variation
in counting
efficiency
that frequently
results
from
uneven
distribution
of
activity
in
the
electro-
deposited
plate and variations
in both distance
and verti-
cal alignment
of the counting
plate
with
respect
to the
detector.
If it is arranged
so that
g, Ez and
t do not
contribute
significantly
to the error,
the absolute
uncer-
tainty in the ~
concentration
in dpm/g equals

where SX indicates
the uncertainty
in X and is taken equal
to (X+
Bx)%. When the blanks
or other
corrections
are
negligible
compared
to
the
total
integral,
SX becomes
equal to X%, and the fractional
error function
reduces to

or the square root of the sum of the reciprocals
of each of
the total
counts
involved, which is simpler to use. If the
quantity
of ~%
and/or
the counting
time used in the
standardization
is sufficiently
large,
the
error
function
simplifies
to the first two terms in either equation.
If the
quantity
of 23%
tracer used in the sample and the yield
are both
sufficiently
high,
the
total
uncertainty
in the
determination
will be determined
entirely
by the uncer-
tainty in the ‘~
count, as it should be.
With
the
small quantity
of 2%
tracer
used
by
many workers,
the uncertainty
in the yield determination
becomes
much greater
than
the uncertainty
in the total
count
of the ‘~
from
the sample.
For example,
if a
100-g
sam le
cent aining
0.1 dpm/g
were
traced
with
3 dpm
of L Pu with
a yield of 50% and the final pluto-
nium fraction
were count ed for 103 min at 25% counting
eftlciency
on a clean detector,
and the same quantity
of
tracer
were standardized
under
the same conditions
but
with a yield of 100%, the overall fractional
error would be

1
1
1
1250
‘%+750

or 0.069.
The resulting
uncertainty
of 13.870 at the 95%
confidence
level is probably
acmptable
in the determina-
tion
of
the
low
levels presently
resulting
from
global
fallout.
However,
it is undesirably
large for more precise
needs at higher
levels and is unnecessary
in any case. At
the
95% confidence
level, the
uncertainty
in the yield
determination
alone is 12.6% compared
to only 5.6% due
23%
count
done.
The uncer-
to the uncertainty
in the
tainty
in the ‘h
count
alone could be further
reduced
to 4% if the yield were also increased
to
100%. As the
concentration
of ‘%% in the sample becomes higher, the
same imprecision
becomes
less acceptable
but the overall
uncertainty
in the final answer is still determined
by the
relatively
larger
uncertainty
in the
yield
determination
resulting
from use of too little
tracer.
In fact, it should
not be difficult
to develop
a procedure
whose
recovery
would
be known
more precisely
than
12.670 without
a
separate
yieid determination.
In our experience,
the pres-
ent procedure
is repralucible
to within 5Y0.
On the other
hand,
if large quantities
of ‘Pu
are
used to improve
the statistics
of the vield determination.
.
other problems
are introduced
that are even more serious
when
the
‘%%
content
is low. Plutonium-236
has two
main alpha rays at 5.769 and 5.722 MeV both
of which
are higher
in energy
than
those of either
‘%
or ~.
Although
the
three
isotopes
can be resolved
easily and
completely
with
current
instrumentaiton,
some
of the
alpha particles
from the higher-energy
~Pu
are scattered
continuously
and
quite
uniformly
through
all
lower

energies to zero. The quantity
scattered
is dependent
not
only
on
the
particular
counting
chamber
used and the
quantity
of absorber
present
but also on the condition
of
the detector
itself.
The percentage
scat tered is relatively
small but
if the total
number
of counts
collected
in the
main peak becomes
very large, the number scattered
into
the lower
channels
represents
a significant
increase over
the
normal
background
of a clean detector.
The conse-
quent
decrease
in both
sensitivity
and precision
for ‘%
soon becomes
the overriding
consideration
and makes the
imprecision
in the yield determination
of secondary
im-
portance.
Furthermore,
‘I%
decays to 23*Uwhich decays
in turn
to
22%% both
of which
lie between
‘%
and
2%,
further
complicating
the resolution
and increasing
the scatter.
Even if freshly purified,
the 2MPu will regrow
its 72-yr daughter
to about
0.5% of the ‘xl%
activity
in
6 months,
necessitating
repeated
unification.
However,
J
the
greatest
drawback
is that
2 Pu generally
contains
both
‘%
and ‘%
in quantities
that are easily detect-
able when
large quantities
of 2MPu and/or
long counting
times are used. As with the scattered
radiation,
the result-
ant increase in background
soon becomes
intolerable
in a
rocedure
for the determination
of low levels of’%
and
L
in the environment.
The %
presently
in use in this laboratory,
after
purification
from 232Uand its daughters,
gives 0.004% of
the total
2%1
integral
~r
charnel
(12.S keV) at lower
energies due to scatter
only. The scatter
plus plutonium
contamination
is 0.0770 of the total
‘l%
integral
in the
‘?u
integral (1O channels),
and 0.7% in the ‘%
integral
(16 channels).
If a combination
of ‘l%
activity, counting
time, and counting
efficiency
are chosen so that 103, 104
or 10s total
counts
are obtained
on both
standard
and
sample,
the statistical
uncertainty
at the 95% confidence
level on the yield determination
alone will be 9,2.8,
and
0.9%, respectively.
If we define
the
detection
limit
as
being the net count that is equal to twice its own standard
deviation
and take 3 pulses in the particular
integral
as a
normal detection
limit on a clean detector,
the increased
background
from
these
same three
levels of total
2MPu
counts
would raise the detection
limit by about
1.7,2.7,
and 7 times, respectively,
for a 10-charnel
integral due to
scattering
only; by about 2, 3.3, and 9 times, respectively,
for the 23%
integral;
and by about
3.3, 9, and 26 times,
respectively,
for the 2%
integral.
The
increased
background
has a similar
effect
in
decreasing
the
precision
of the
determination
and
the
uncertainty
increases
either
as the
sample
activity
de-
creases or as the quantity
of 2%% used increases.
Conse-
quently,
a compromise
is necessary,
and the quantity
of
tracer used should be much less for low-level samples than
for high-level ones. Because the concentration
of the 2%J
tracer
is the fundamental
value on which all subsequent
analyses
depend,
its determination
should be carried out
as carefully
and accurately
as possible,
using as least as
many total counts
as will be obtained
subsequently
from
the highest sample to be analyzed.
The standardization
is
completely
separate
from
any actual
sample
analyses so

.

.

.

52

.

.

that
large numbers
of counts
can be used without
prob-
lems due to scatter
or contamination
with
other
pluto-
nium
nuclides.
In fact,
a large count
will be helpful
in
determining
the scatter
and contamination
with adequate
precision.
Consequently,
the uncertainty
in the determin-
ation
will depend
entirely
on the number
of counts
of
‘%%
and ~
obtained
in the analysis.
If the yield is
also high, even the *
count will not contribute
signifi-
cantly
to the imprecision
until the 23%
ccmnt becomes
nearly
equal.
For example,
in this laboratory,
a totaI of
104 to
10s
counts
are used for standardization
of the
‘l%
tracer
and
determination
of the
scatter;
2 x 103

counts
are used on background-level
samples up to about
0.8 dprn/g
using a 103-rein
count
at 25% counting
effi-
ciency
on
a
10-g
sample;
104
counts
are
used
for
medium-level
work up to about
4 dpm/g; and 10s counts

are used for highest precision on higher levels at which the
increased
scatter
and contamination
will be relatively
in-
significant.
The upper
end of the two lower ranges is the
level at which the uncertainty
in the ‘%% count becomes
equal
to
that
in the yield determination,
i.e., the total
counts of’%
and *
recovered
are equal.

Reference

1.
R. J. Overman and H. M.Clark, “Ra&;oisotopeTecW~ques;’
McGrawHill, New York, N. Y., 1960, p. 109.

.

53

EXPERIENCE
GAlNED
FROM
AN EXTENSIVE

OPERATIONAL
EVALUATION
OF THE
FIDLER

b

D. R. Case, W. T. Bartlett, and G. S. Kush

USAF
Radiological Health Laboratory

Wright-Patterson AFB,
Ohio

ABSTRACT

The prompt
assessmentof plutonium
distribution
resulting from nuclear

weapons eccidant/incident
debris depends stongly on the ability to deploy an

operational y
reedy
team
of
thoroughly
tiained
personnel
equipped
with

reliable equipment.
A program of routine testing of four
FIDLER
response kks

has resulted in a complate
characterization
of the instrument and a comple-

ment
of
personnel
a~uainted
with
its operation,
shortcomings,
and,
field

application.
Results of statistical reliability
tests on the FIDLER,
a discussion

of instrumental dafkiencias
observed, and a summary of an accident/incident

training program will ba presented. The ex~rience
gained from such a program

allows the USAF
Radiological Health Laboratory
to fulfill
its responsibility for
worldwide Air Force weapon accident/incident
hazard evaluation.

-.-.
—..
-.. -.-.
-.. -.---
...
..........
... .. ..

Introduction

The
ability
to
promptly
evaluate
the radiological
hazards associated
with nuclear weapons accidents
and/or
incidents
is of prime interest
to the Air Force. To satisfy
this requirement,
the USAF Radiological
Health
Labora-
tory
has been
tasked
with
providing
an immediate
re-
sponse capability
in the event of such an occurrence
on a
worldwide
basis. We have prepared
for this task by insti-
tuting
a program
for acquiring
and maintaining
appropri-
ate instrumentation,
and
for
training
a complement
of
personnel
in the use of this instrumentation
in evaluating
the distribution
of accident/incident
debris. This program
has, as two prime objectives,
the familiarization
of person-
nel with
the
actual
equipment
and the maintenance
of
equipment
in an operationally
ready
status.
The
basic
equipment
employed
for the detection
of plutonium
and
daughters
is the
Radiac
Set P/N 400520,
whose primary
component
is the FIDLER,l
a scintillation
instrument
for
detection
of low-energy photons.
The basic characteristics
of this instrument
have been outlined,l-3
as well as investi-
gations
on the temperature
dependence,4
and effects of
overburden.5
These
investigations
have
served
well to
supply
the
basic
characteristics
of
the
instrument.
In

order
to incorporate
the FIDLER
into a response
ready
program.
Additional
information
was necessary
to evalu-
ate
its
serviceability
and
to
identify
and remedy
and
deficiencies
in its longterm
reliability.
A program
for
routine
calibration
and evaluation
of the stability
of the
FIDLER,
coupled
with field training sessions for response
persomel,
has been carried out for a period of 14 months.
Evaluation
of the statistical
reliability
of the instrument
has aided
in the identification
and correction
of several
problem
areas which could have hindered
the validity
of
the
FIDLER
in a field
situation.
The
result
of such a
program
of
testing
and
training
is to
insure
that
the
instrumentation
will be operational
when needed,
and to
provide
thorough
familiarization
with the equipment
for
those using it.

Methodology

The
Radiac
Set P/N 400520
(Eberline
Instrument
Co.)
consists
of
three
probes
(a FIDLER
scintillation
probe,
a PC-2 scintillation’
probe,
and a SPA-3 scintilla-
tion probe),
a PRM-5 pulse-rate
meter,
and various acces-
sory components,
housed
in an aluminum,
flex-hair-lined

55

carrying
case. The PRM-5 is a battery
operated
rate meter
with
pulse-height
analysis
capability
and
supplies
three
switch-select able,
independent
ly
adjustable
high-voltage
settings.
A total of four kits were employed
in this study.
For
routine
use in Broken
Arrow operations,
the
PRM-5 is set up to provide maximum
response
to pluto-
nium
and its daughters.
Generally,
the pulse-height
anal-
yzer is operated
with
a 100% window
width.
The three,
switch-selectable
high voltages are adjusted
as described
in
Table I.
The long-term
testing
of the instrument
reliability
consists
of performing
measurements
of the response
of
each
instrument
to the
17 keV and 60 keV photons
of
‘lAm.
Since these respnse
checks are incorporated
into
familiarization
sessions, two procedures
are followed.
The
first check consists
of measuring the response of both the
FIDLER
and
PC-2
probes
to a nominal
100 nCi ‘lAm
source
in contact
with the detector
face. Net counts
per
minute
are tabulated
and used to calculate running means
and standard
deviations.
The second portion
of the testing
procedure
consists
of a calibration
of the point and area
sensitivity
of each probe
using procedures
described
by
Tlmey.6
Each
detector
is suspended
at
a height
of
30.5 cm (12 in.) above a surface and the response
of the
instrument
to a 9.82 gCi ‘lAm
point
source is measured
at O, 5, 15, up to 105 cm. Point and area source sensitivi-
ties are calculated
according
to the

Sp (cpm/#Ci)
= *

Sa(cpm/flCi
o m’)=
2 ‘~o-’n

following equations:

(1)

X (R)(N)
(2)

where

Sp
=
point source sensitivity
Sa
=
area source sensitivity
Q
=
source strength
in uCi
R
=
radial distan-m of each response in cm
N
=
response at radial distance
R

These data are also tabulated
and used to calculate a mean
and standard
deviation for each instrument.
Data for each
session
are compared
to the average and used to deter-
mine the need fo; corrective
action.-

TABLE I

HIGH VOLTAGE SE’ITINGS

Switch
Position
Probe

HV1
FIDLER
HV2
FIDLER
HV3
PG2

Energy
(kev)

17
60
17

Results and Discussion

The data accumulated
over a period extending
from
22 April 1970 through
30 June 1971 have been summar-
ized and are shown in Table II. Mean and standard
devia-
tion
values
are shown
for point
source
sensitivity
(Sp),
area source
sensitivityy (Sa), and check
source
response.
These results
indicate
that over a long-term
period,
both
the FIDLER
and PG2
are reproducible
to within a 10 to
15% range. This correspondence
is achieved with a mini-
mum of preventive
maintenance
or attempts
to ccmtinual-
Iy optimize
the
settings
of the
instruments.
In fact,
a
comparison
of
individual
data
with
the
averages
has
proven
to be of value in detecting
instrument
deficiencies
such as maladjusted
high-voltage
settings,
incorrect
win-
dow widths, and malfunctioning
multiplier
phototubes.
In addition
to in-house
maintenance
of this equip-
ment,
we
provide
assistance
to
other
Air
Force
and
Government
agencies on the operation
of the Radiac Set.
One particular
problem
has arisen in obtaining
adequate
response
of the FIDLER
probe
to 17 keV photons.
Ad-
justment
of the high voltage
to satisfactorily
center
the
17 keV
peak
has been
encountered.
Through
a careful
study
of the correspondence
of high voltage
applied
to
center
a given photopeak
in the window,
we have deter-
mine-d that the 17 keV peak position for the FIDLER
and
the
maximum
output
of the
PRM-5
are both
approxi-
mately
the same (1370 V). The difficulty
has been cor-
rected
through
modifications
to
the
power
supp!y
to
allow a maximum
output
of 1600 V. This increased volt-
age allows a more careful adjustment
of the 17 keV peak
in the analyzer window.
An exhaustive
program
for
training
of
response
persomel
has also been instituted.
This training
consists
of
in-house
efforts
to
provide
realistic
situations
and
periodic
deployment
of the equipment
and personnel
in
aid of actual and/or
anticipated
radiological
hazards. Our
in-house
training
consists of sessions conducted
by a staff
of Health Physicists to acquaint
personnel
with the theory
of operation,
calibration
and set-up,
and field use prob-
lems of major significance
to the successful
utilization
of
the kit.
Field exercises
are also utilized
to provide
prac-
tical experience
under simulated
plutonium
distributions.
The effects of overburden,
response time, etc., are demon-
strated
and
coupled
with
instruction
in proper
survey
techniques.
In addition,
persomel
have
been
deployed
with
the
Radiac
Sets to aid in the evaluation
of existing
contamination
areas. These
teams
have also aided in the
health
physics
support
of
Apollo
shots.
These
deploy-
ments
are considered
of great
value
in complementing
in-house
training
and in providing
continual
reevaluation
of equipment
and techniques.
It should
be pointed
out
that the
Radiac
Set has been
found
to be a very easily
deployable
instrument.

.

.

56

TABLE II

RESPONSE
DATA

Function

FIDLER HV-I (Sa)
FIDLER HV-2 (Sa)
PG-2 HV-3 (Sa)
FIDLER HV-1 (Sp)
FIDLER HV-2 (Sp)
PG-213V-3 (Sp)
FIDLER Check HV-1
FIDLER Check HV-2
PG-2 Check HV-3

X = mean value
o = one standard
deviation

unit
1

%
u

2906
212
3328
442
148
19
5053
61
4937
141
606
85
28.4K
6.OK
33.3K
8.2K
9.7K
0.9K

Unit 2

2613
3252
157
7010
7447
574
23.6K
29.2K
6.74K

Summary

This program
for priodic
evaluation
of the Radiac
Set coupled
with a program
of training
for personnel
has
allowed
this laboratory
to achieve an operationally
ready
status.
The testing program
has provided
a basis for con-
tinually
assuring
that
our
equipment
is operating
in a
reliable
manner.
In addition,
necessary
modifications
to
improve
the
reliability
of the Radiac
Set have been
in-
corporated
as a result
of this testing.
These experiences
have
allowed
us to
gain
confidence
in our
ability
to
promptly
respond
to the need for radiological
assessment
of any situation
involving fissionable
materials.

References

1.

2.

C. T. Schmidt and J. J. Koch, “Plutonium Survey and X-Ray
Detectors:
in Hazards Control
Progress Reprt
No.
26,

Lawrence
Radiation
Laboratory,
Livermore,
Rept.
UCRL-50007-66-2(1966), p. 1.

J. F. Thmey and J. J. Koch, “An X-Ray Survey Meter for
Plutonium Contamination, “ in Hazards Control Bogress Re-
port No. 29, Lawrence Radiation Laboratory, Livermore, Rept.
UCRL-50007-67-3(1967) p. 6.

140
348
54
858
709
89
5.3K
6.7K
2.5K

Unit 3

i
u

1786
170
3195
188
158
38
5082
36
5009
105
628
144
23.4K
4.2K
27.8K
5.5K
7.lK
2.5K

Unit 4

2527
2837
263
5645
5851
731
23.2K
26.OK
9.lK

o

250
521
88
766
729
191
5.OK
5.9K
4.OK

3. C. L. Lhdekin and J. J. Koch, “Optimization Studies for the
FIDLER
Detector;’
in Hazards
Control
Report
No.
31,

Lawrence Radiation Laboratory,
Livermore, Rept.
UCRL-
50007-68-2 (1968) p. 20.

4. T. O. Hocger and J. F. Tinne~, “Temperature Dependence of a
Plutonium X-Ray Survey Instzument~ in Hazards Control Re-
mrt No. 33. Lawrence Radiation Laboratozv. Livermore. Reut.
tiCRL-50007%9-l (1969) P. 14.
-.

5.

6.

J. F. Thmey and T. O. Hoeger, “Overburden Attenuation
mpU-241Am us~g the FII)LER Detector!”
Measurements for
in Hazards Control Report No. 33, Lawrence Radiation Labor-
atory, Livermore, Rept. UCRL-50007+9-I (1969) p. 6.

J. F. Tinney, “calibration
of an X-Ray Sensitive Plutonium
Detector;’ “m Hazards Control Report No. 31, Lawrence Radia-
tion Laboratory, Livermore, Rept. UCRL-50007-68-2 (1968)
p. 24,

57

.

SEPARATION
AND
ANALYSIS
OF PLUTONIUM
IN SOI L

by

G. E. Bentley, W. R. Daniels, G. W. Knobeloch,

F. O. Lawrence, and D. C. Hoffman

Los Alarms Scientific Laboratory

University of California

Los Alamos, Naw Mexico

ABSTRACT

A
procedure for
the analysis of
plutonium
in large samples of soil has

been developed which gives plutonium
yields of at least 90°L. The soil samples

are completely
dissolved by repeated f umings with
HNOa,
HF
and
HC104,

followed
by treatment with NaOH
to give silicate-free solutions in either HCI
\
or HN03.
As much as 50 g of soil, with final concentrations corresponding to
~ 100 mg/ml of solution, have been dissolved. (Processing of larger amounts of

material appears to be limited only
by the volume of
solution
that can be

handled.)
The
sample may
be traced
by
adding
an appropriate
plutonium
isotope.
NaN02
is added
to
insure
that
all
of
the
plutonium
is in the

(lV)-oxidation
state, thus
providing
for
exchange between
the
plutonium

tracer and the plutonium
in the sample. The solution is extractad into di-2-

ethylhexyl
orthophosphoric
acid (HDEHP);
the HDEHP
is then washed several

times with
6h4 HCI
to
remove
iron.
After
the washing, 2,5-cJitertiarybutyl-

hydroquinone
(DBHQ)
is added to reduce the Pu(IV)
to Pu(lll),
which is then

back-extracted into
6~
HCI. The plutonium
may then be determined
by any

standard method.

..--. -.. --. -.. -.----
.-. - ..........-.. -—-------

Introduction

In connection
with the responsibility
of the LASL
Radiochemistry
Group
for
the
analysis
of
the
under-
ground
debris resulting from the testing of nuclear devices
at the Nevada Test Site, procedures
for the quantitative
analysis
of
plutonium
in
soil
utilizing
extraction
into
di-2-ethylhexyl
orthophosphoric
acid (HDEHP) have been
developed.
Procedures
involving
coprecipitation
with
LaF3 are not suitable
when large volumes of solutions
of
high ionic strength
are to be analyzed.
The present
work
describes
the adaptation
of our standard
procedures
for
the dissolution
of dirt and the extraction
of plutonium
to
the separation
of low-level plutonium
from surface soils.

Experimental Method

When plutonium
is to be determined
in soils con-
taining
no detectable
activity
with which
to follow
the

yield of various steps in the dissolution
procedure,
quanti-
tative
recovery
of
plutonium
is insured
by
completely
dissolving the soil sample by fuming with HF, HN03
and
HCIOq, followed
by treatment
with NaOH and then HC1.
The plutonium,
in either the (IV)- or (VI)-oxidation
state,
can
then
be extracted
into
HDEHP
in gheptane
from
HN03
or HC1 solutions
of a wide range of concentrations.
We have found
6~ HCI solutions
to be convenient
since
the extraction
coefficients
for iron and many other
con-
taminants
show minima
at this molarit y. However,
since
the extraction
coefficient
for I%(IV) in 6~ HC1 is about
an order of magnitude
higherl
than for Pu(VI), NaN02
is
added
to insure
that
the plutonium
is in the (IV) state.
(This also provides
for exchange
if plutonium
tracer has
been added.)
The plutonium
is recovered
from the extract ant by
addition
of
2,5-ditertiarybutylhydroquinone
(DBHQ)
which reduces
the plutonium
to the (III) state and strong-
ly complexes
it. The I%(III) may then be readily removed
from the organic
phase by extraction
with dilute HC1. At

59

this
point,
the
bulk
of the soil components
have been
removed
since monovalent
and divalent
species,
such as
sodium
and calcium,
and most
trivalent
spcies
will not
have been ext ratted
into HDEHP under these conditions.
Further,
most of the higher oxidation
state species (e.g.,
zirconium),
which
have been extracted
will not be back-
extracted.
Large amounts
of iron, which
interfere
with
the subsequent
plutonium
analysis, can be eliminated
by
performing
the
initial
extraction
from
6hj HC1 and
by
washing with 6~ HC1as required.
The
final
solution
containing
the
back-extracted
plutonium
can not be concentrated
and analyzed
by any
standard
met hod.2
Since
the
initial
soil dissolution
is
quantitative
and yields of 90% can be achieved through
the
extraction,
the
sensitivity
of the method
is limited
only
by the
amounts
of soil dissolved,
the volumes
of
solution
one
wishes
to
handle
at
one
time,
and
the
a-counting
system
to be used. The procedure
has been
applied
to samples containing
as little as a disintegration
per minute of plutonium
activity.

Experimental
Procedure

Dksolution
of Soil Samples.
An =50
g sample
of
the pulverized
soil is placed in a Teflon beaker and 50 ml
of fuming
HN03
is added.
The
mixture
is slurried
by
stirring
with a stainless
steel stirring
rod until all of the
dry
powder
is thoroughly
wet.
100 ml of concentrated
HCIOQ is added to the slurry, and this is followed by the
gradual addition
of 100 ml of concentrated
HF. The addi-
tion of HF is accompanied
by the release of voluminous
quantities
of gas. The mixture
must be cooled in a water
bath
and the HF added in small portions
to prevent
the
solution
from
overflowing
the beaker.
The effervescence
subsides
appreciably
after
= 75% of the
HF has been
added.
After
addition
of
the
HF,
the
Teflon
beaker
is
heated on a hot plate (medium
setting),
to heavy fumes of
HCIOQ. The beaker is cooled in a water bath,
and 50 ml
of HF is added.
(If the beaker
is not sufficiently
cooled,
the HF will spatter rather violently when it is added.) This
HF
fuming
step is performed
three
more
times,
adding
HCIOA if necessary to prevent the mixture
from becoming
completely
dry. During the fourth
fuming the contents
of
the
beaker
are taken
almost
to
dryness.
The beaker
is
cooled
and
100 ml of 4&l HC1 is added.
The mixture
is
boiled.
The
contents
of the
beaker
are transferred
to
40-ml short-taper
Vycor centrifuge
tubes and centrifuged.
The supemate
is poured
into a second Teflon beaker and

50 ml of
concentrated
HF and
50 ml of concentrated
HCIOQ are added.
The beaker
is then
heated
on a hot
plate (medium
setting).
The original beaker is rinsed with
hot 4~ HC1, and the wash is transferred
to the centrifuge
tubes
cent aining residue.
The contents
of the tubes
are
stirred
and cent rifuged,
and the supemates
are added to
the second
beaker.
Bach tube containing
residue is boiled
over
a
burner
with
x 2 ml
of
6hj NaOH.
Sufficient

4M HC1 is added
to acidify
the mixture.
The solution
is
~–tin boiled and centrifuged
while still hot. The supernate
in each case is added
to the second
Teflon beaker.
The
treatment
with NaOH and HC1is repeated,
and the super-
nates are again added
to the second beaker.
The residues
in the tube are transferred
to the original beaker with HC1
and treated with four HF-HC104 fumings.
The
contents
of
the second
beaker
are heated
to
heavy
fumes
of HCIOq
and
cooled.
Fifty ml of
HF is
added
to
the
solution
which
is then
fumed
almost
to
dryness
and again cooled.
Then
= 100 ml of 6~ HCl is
added.
The mixture
is warmed,
transferred
to Vycor cen-
trifuge
tubes and centrifuged.
The supernates
are poured
into
a polyethylene
bottle.
Any remaining
residue is re-
peatedly
boiled
with 6~ HCI, centrifuged
and the super-
nate is added to the polyethylene
bottle.
The HC1dissohr-
tion
treatment
is continued
until no visible reduction
in
the amount
of residue is observed.
The contents
of the original
beaker
are fumed
al-
most
to dryness,
and = 100 ml of 4~ HC1 is added.
The
mixture
is boiled and transferred
to the centrifuge
tubes
containing
the insoluble
residue from the second beaker.
The
contents
of the
tubes
are stirred
and centrifuged.
Again, the supernates
are poured
into the second beaker
and fumed
twice with HF-HC104.
Any precipitate
in the
tubes
is treated
with
NaOH-HCl
as described
previously
and the mixture
centrifuged.
The supernates
are added to
the
second
beaker.
Then,
if any residue
remains
in the
centrifuge
tubes,
HF-HC104
fumings
are re~ated
until
NaOH-HCl treatment
gives complete
solution.
The result-
ing solutions
are added to the second beaker. The solution
in the second beaker is treated with 50 ml each of concen-
trated
HF and HCIOq, taken
to heavy fumes of HC104,
and cooled.
Then 50 ml of concentrated
HF is added and
the solution
is fumed
almost
to dryness.
The residue
is
dissolved
in 6~ HC1 and
the solution
is added
to the
polyethylene
bottle.
The final solution
tends to salt out on standing for
several
days.
However,
heating
of the
solution
just
to
boiling causes the precipitated
salts to redissolve.

Plutonium
Extraction.
A suitable
plutonium
tracer,
usually
2*Pu,
is added
to the sample
solution
for yield
determination.
Suftlcient
10~ NaN02
is added to make
the solution
0.2M in this reagent. The resulting solution
is
heated just to b~iling and cooled to room temperature.
A
volume
of
1~ HDEHP
in g-heptane
equivalent
to one-
third
that
of the sample is pre-equilibrated
with 6~ HC1
and
added
to the
sample
in a separator
funnel.
The
mixture
is shaken for 1 rein, and the organic (upper)
and
aqueous
phases
are
allowed
to
separate.
The
aqueous
phase is discarded.
The organic layer is washed five times
with equal volumes of 6M HCI and the washes discarded.
The
HDEHP
solution
is shaken
for * 10 sec with
one-
third its volume of 0.2M DBHQ in 2*thyl-1 -hexanol. The
plutonium
in the resul~ng
mixture
is back-extracted
by
shaking
for 2 min with
one-half volume of 6~ HC1. The
phases are allowed
to separate
for 5 min and the organic

.

.

60

layer
is discarded.
The aqueous
solution
is reduced
in
volume
to 5 ml or less by boiling and water is added to
make the solution
3~ in HC1,with the final volume being
no more than 10 ml.
The final plutonium
separation
and determination
are carried
out by a standard
LaF3
coprecipitation
fol-
lowed
by an anion
exchange
resin
column
technique2
involving
elution
of
the
plutonium
from
the
resin
by
reduction
of Pu(IV)
to the
(III)
state
with
an HI-HC1
mixture.

Discussion

Samples
of surface
soil were
collected
from
five
locations
at the Nevada Test Site. The samples were taken
from
areas
which
were
believed
to contain
little
or no
plutonium.
About
500 g of dirt (avoiding rocks>
2 cm in
diam)
was obtained
from
the
surface
at each sampling
point.
No activit y could be detected
in any of the samples
with an alpha-survey
meter.
Two S=50 g portions
of each dirt sample were dis-
solved,
giving
final
concentrations
corresponding
to
* 100 mg of soil per ml of solution.
The plutonium
was
extracted
by the described
HDEHP procedure.
No diffi-
culties were encountered,
and, in fact, the high dirt con-
centration
seems to aid the phase separation
during the
initial
extraction.
A SO-MI aliquot
of solution
from each
sample was analyzed
without
adding plutonium
tracer
so
that
any
isotopes
of plutonium
present
in the
sample
could be determined,
and the appropriate
choice of tracer
made.

Because
of the
time
(~ 20 h) that
is required
to
dissolve the samples using this procedure,
it would not be
practical
to
use
it
to determine
plutonium
in a large
number
of samples.
The procedure
would,
however,
be
useful to check a faster leach-type
of procedure
for com-
pleteness
of plutonium
recovery.
This is especially true if
samples with very low amounts
of plutonium
were being
determined.
Our procedure
could
be shortened
considerably
if
the small amount
of sand-like residue remaining
after one
complete
cycle could be discarded.
In the application
of
this procedure
to debris from
nuclear
devices,
the large
amount
of gamma activity provides a measurement
of the
completeness
of dissolution;
inactive residues
may be dis-
carded.
Possible
future
work
might
involve
the
use of
tracers
to
determine
the advisability
of discarding
such
residues.

References

1. ORNL Chemical Technology Division, Annual ProgressReport
ORNL4272, May, 1968.

2. J. IOeinberg (cd.), Cotlected Radiochemical Procedures, Los
Alamos Scientific Laboratory
Report
LA-1721, 3rd Ed.
(1967), p. 91.

61

COMPARISON
OF A LEACHING
METHOD
AND
A FUSION
METHOD

FOR THE
DETERMINATION
OF PLUTONIUM-238
IN SOI L

by

C. T: Bishop, W. E. Sheehan, R. K. Gillette, and B. Robinson

Monsanto Research Corporation

Mound
Latmratory

Miamisburg, Ohio

ABSTRACT

Both a leaching and a fusion procedure, followed
by alpha pulse-height

analysis, were used to determine the plutonium
content of four
soil samples.

Thirty-one
plutonium
determinations
were made following
an acid leach pro-

cedure. Twentyane
plutonium
determinations
of these same four soil samples

were mada following
the
potassium fluoride-pyrosulfate
fusion
method
de-

veloped by C.W. Sill and K. W. Puphal. Plutonium
concentrations
in the four

soil samples analyzed
were found
to be 0.04,
0.19,
1.6, and 20 dis/min
of
Z6pu/9 of soil. Leaching and fusion results were essentially in areement.
AS a

further
check, eight leached residues from
one of the four
soil samples were

dissolved by the fusion
method
and analyzed;
results indicated that greater

than 90% of the ‘*u
was removed from the soil by acid leaching.

Comparison
of the precision of the fusion procedure with the precision

of the restdts of the four soil samples analyzed by the fusion method indkates

a nonuniform
distribution
of plutonium
in the soil. This is probably
due to the

particulate nature of the plutonium
contaminants in the soil.

Introduction

Early in 1970, Mound
Laboratory
initiated
a pro-
gram to develop
an improved,
relatively
simple and reli-
able analytical
procedure
for the routine
determination
of
plutonium
in
soil. Prior
to July
1970,
all soil sample
analyses had been performed
by the Environmental
Con-
trol Analytical
Group using an acid-leach
method
of dis-
solving
the
plutonium
from
the
soil. By July
1970,
a
serious debate
was well under way in the scientific
com-
munity
concerning
the effectiveness
of the leach method
as compared
to a total
dissolution
of the
soil accom-
plished by conventional
fusion methods.
To
evaluate
these
two
methods
of
plutonium
dissolution
from soil and achieve our own assurance
that
methods
being
used
at Mound
Laboratory
for
routine
plutonium
soil analyses were reliable,
the Analytical
Sec-
tion
of the
Nuclear
Operations
Department
performed
analyses
on a select number
of soil samples by a fusion

procedure.
The
four
soil
samples
used
in
this
study
covered
a wide range of -u
concentration,
i.e., from
0.04 dis/min/g
to 20 dis/min/g
of “%.
It is significant
that
these four soil samples were analyzed
by two essen-
tially independent
analytical
laboratories.
The personnel,
counting
systems,
and standards
employed
in the fusion
determination
were all different
from those employed
in
the
leaching
method.
The
purpose
of this
report
is to
present
the results of the analyses of these four samples,
and
to
show
the indicated
agreement
between
leaching
and fusion methods
in the determination
of %
in soil.
The composite
soil samplel
is dried in stainless steel
pans on a hot
plate.
The core samples are placed in the
pans in such a manner
that
the vegetation
on the surface
of the individual cores can be charred by a propane
torch.
After the vegetation
is charred
and the soil aggregates are
broken
up, the sample is mixed well for complete
drying.
The
samples
are ground
with
a mortar
and pestle.
The
larger rocks, those not passing through
a 20-mesh screen,

63

are removed
from the sample.
The remaining
sample
is
ground
and
screened
through
a 35-mesh
screen,
placed
into
a one-gal
plastic
container,
and
weighed.
Fifty-g
aliquots
are weighed and analyzed
by one of the two acid
leach
methods.
Teng
aliquots
are
used
in the
fusion
analyses.

Acid Leach Method

The flow diagram
in Figure
1 summarizes
the two
acid
leach
procedures
that
have
been
used
at Mound
Laboratory.
On the left side is the original procedure
by
which the leach results reported
here were obtained.
The
procedure
currently
in use (referred
to
as the
current
method)
is shown on the right side of Figure 1.
In the original method,
the -u
tracer is added to
the soil and the satnple
is placed
in a muffle
furnace
at
500”C for 30 min to convert the ‘I%
tracer to an oxide.
This
sample
is then
leached
by
vigorous
shaking
for
approximately
1 h with
100 ml of concentrated
nitric
acid and 1 ml of concentrated
hydrofluoric
acid at room
temperature.
After
standing
overnight,
the
solution
is
separated
from the soil and adjusted
to 4N in nitric acid.
The plutonium
is extracted
into
a 10% ~isooctylamine
(TIOA)-xylene
solution according to the method
reported
by F. E. Butler.z
The plutonium
is back-extiacted
from

ME inal
Method
9= rent
Nethod

I?%!%ht“’-’’’l”1

current ndlod
orl@lAl
NethOd
~------
---------
----.
-------
.—---
_
: AVR.recovery 8~ h 192 ~
~ Ave.recOV=# U% * zn ~
I No. of •tul~se~ Z8
----------------
J
o No. of -nal~ses 31
0
L.------
.- .-------J

~ud.=ov.ry
is bssed @Z
m tho.a
muly..s
performed fox this

Fig. 1

the TIOA-xylene
solution
with dilute nitric acid contain-
ing sulfur
dioxide.
This
solution
is adjusted
to 10N in
hydrochloric
acid,
passed
through
a chloride
anion
ex-
change
column,
and
eluted
with
6~ hydrochloric
acid
containing
!).024% hydrogen
iodide
according
to
the
method
reported
by L. C. Henley.3 The eluted solution is
taken to dryness in nitric acid, and an ammonium
sulfate
electrolytic
plating
bath
is prepared
according
to the
method
reported
by I. A. Dupzyk.4
The
current
leach
procedure
closely
follows
the
method
reported
by N. Y. Chu.s
In this method
100 ml
of a 3-to-1, by volume,
mixture
of concentrated
nitric to
concentrated
hydrochloric
acid is used to leach the pluto-
nium
from
the
soil. Here
the
mixture
is heated
while
stirring
for
1 h at near
boiling
temperature.
The leach
solution
is removed
and a second
leach is carried out in
the same way.
Both leach solutions
and a water rinse of
the
soil residue
are combined
for further
analysis.
This
solution
is evaporated
to
near
dryness
to remove
the .
hydrochloric
acid
and
adjusted
to 7.5~
in nitric
acid.
Sodium
nitrite
is then added
to the solution
to ensure a
+4 oxidation
state
for the plutonium
before
it is passed
through
a nitrate anion exchange columns
The column is
rinsed
with
concentrated
hydrochloric
acid
as the first
measure
to separate
the natural
thorium
from the sample.
The plutonium
is then eluted with 61j hydrochloric
acid,
con-taining 0.024% hydrogen
iodide. The elut ed solution
from
the
nitrate
column
is adjusted
to
10N in hydro-
chloric
acid,
passed
through
a chloride
anion
exchange
column
as
a
final
decontamination
step
for
natural
thorium,
and
finally
eluted
and electroplated
as in the
original
method.
The
complete
decontamination
of
natural
22%
is essential
for a‘%
determination
due to
the closeness of the minor “%
alpha energy (5 .46 MeV)
and the maximum
2%
alpha energy (5.42 MeV).
In summary,
the
changes
in the
leach
procedure
were replacement
of the nitric acid leach with the method
reported
by Chu,
and substitution
of the nitrate
anion
exchange
column
for the
TIOA
liquid
extraction
step.
The improvement
gained by the current
leach procedure
is that
metals
such as iron and lead that
interfere
with
electrode posit ion are more completely
separated
by the
nitrate
anion
exchange
column.
This
results
in better
recoveries
of plutonium,
and reduces slide deposits during
electrodeposition
to produce
a much better
alpha source
for more effective
pulse height analysis. Tracer recoveries
using the original
procedure
were quite
low and erratic,
46 * 27%, while
the
recoveries
using the modified
pro-
ce d ure
have
been
generaUy
much
higher,
namely
82*
19%.

Fusion Method

The fusion method
used in this study
is essentially
identical
to the method
developed
by C. W. Sill, et al.G-8
A summary
of the procedure
is given in Figure
2. This
procedure
involves
fusing
the
soil
with
anhydrous

.

.

64

.

.

#
Anhydrous
Removal
of
Second

GKF
Fusion
-
Si
& HF
with
4
Fusion
with
2=6Pu
Tracer
HQS04
(Pyrosulfate)
J
I

)
t

Dissolution
Coprecipitation
Dissolution
Extraction
into
of-Pu-238
d
of
Aliquat-336
with
BaS04
Melt

t
b
G

Pu-238
Electro-
Alpha
Back
D
deposition
.
Pulse
Height

1 Extraction
(Oxalate-Chloride)
Analysis
,

Fig. 2

potassium
fluoride
followed
by a pyrosulfate
fusion
to
completely
decompose
the
soil. The
solidified
melt
is
dissolved with a potassium
metabisulfate
solutiorr andthe
plutonium
isseparated
from thesolution
bycoprecipita-
tion with barium
sulfate.
’fhebariumsulfate
is dissolved
in an aluminum
nitrate
solution
and
the
plutonium
is
extracted
into
Aliquat
336
(General
Mills,
Inc.,
Minneapolis,
Minnesota)
nitrate
in
xyiene.
Interfering
metals are removed
by back extraction
before
the pluto-
nium
is back
extracted
with
an oxalic
perchloric
acid
stripping
solution.
After
evaporation
to dryness and dis-
solution
of the residue
in a mixed oxalate<hloride
elec-
troIyte,
the
plutonium
is
electrodeposited
by
the
procedure
developed
by K. W. Puphaland
D. R.01sen.9
The plutonium
is finaUy determined
by alpha pulse-height
analysis
utilizing
a
4096
multichannel
analyzer
and
300 mm2 surface barrier detector.

The -u
tracer
in the fusion procedure
indicated
less than 8070 recovery
of plutonium.
Tracer studies indi-
cated greater
than 959Z0recovery from the initial fusion of
the soil, through
the coprecipitation,
the solvent extrac-
tion,
and the preparation
for electrodeposition.
Electro-
deposition
efficiencies,
however,
were
frequently
much
less than 95%. For this reason 2%
tracer was used in all
analyses of soil by the fusion procedure.
To evaluate the accuracy
and precision of the fusion
procedure
two stancimi
plutonium
soil samples were an-
alyzed.
One sample
was prepared
at Mound
Laboratory
by spiking a soiI sample with a standard
soIution of 2~u,
and
the
other
was a soil sample
spiked
with
‘%
ob-
tained from C. W. SiU. The results of the fusion analyses
are given in Tables I and II. In both
samples,
the experi-
mental
average
agreed
to
within
a few percent
of the
standard
value.
The
relative
standard
deviation
of the

TABLE I

ANALYSIS
OF A 238Pu“STANDARD
SOIL SAMPLE BY THE FUSION METHOD

Weight of
‘EPu in Sample
23%%in Sample
Sample
Sample
(Standard
Value)
(Found)
Number
k)
(dis/min/g)
(dis/min/g)

A-1
1
36.4
36.0 k 1.6’
A-2
1
36.4
34.9 t 1.7
A-3
1
36.4
41.2 ?3.3
A-4
1
36.4
35.2 * 1.6
A-5
1
36.4
35.4?
1.4
A-6
1
36.4
38.0 I 1.4
s-5 1
20
36.4
35.3 * 2.3

Average
36.6 ~ 2.3b
——..——-——
—.——
—-—--——
——

%5tandarddeviation based on counting statistics.

2MPu Tracer
Recovered
(%)

58
45
15.9
18.6
75
65
75

bExperimental
standard deviation
based on the seven individual
determinations.

65

TABLE II

ANALYSIS OF A 23%
SPIKED
SOIL SAMPLEa BY THE FUSION METHOD

.

.

Weight of
‘k
in Sample
23%s in Sample
236PuTracer
Sample
Sample
(Standard
Value)
(Found)
Recovered
Number
(8)
(dis/min/g)
(dis/min/g)
(%)

1
1
35.42
33.48 * 1.67b
51
2
1
35.42
35.26 * 1.49
57

Average
34.37

!Standard soil sample supplied to Mound
Laboratory
by C. W. Sill, Health Services
Laboratory,
U. S. Atomic
Energy
Commission,
Idaho, Falls, Idaho.

bStandard
deviation
based on counting
statistics.

seven ~apu
standard
samples
was 6.3Y0. AS will be seen
later this variation
is low compared
to the standard
devia-
tion observed with actual soil samples.

Control Analyses

Blank
determinations
were
made
periodically
to
examine
the
possibility
of contamination
from
the rea-
gents or glassware. In some cases, a 2%Pu tracer was added
to determine
the percent recovery when a blank value was
determined.
With both
the leaching and the fusion tech-
niques,
the low blank
was about
0.01 dis/min
of ‘~u.
High blank values of 0.09 dis/min
2%1
and 0.20 dis/min
-u
were observed
for leaching and fusion, respectively.
The
average
of
12 leaching
blanks
was 0.036 dis/min
23 Spu,
while
the
average
fusion
blank
value
was
0.070 dis/min
%%I
for 13 determinations.
For most of
the
samples
described
in this report,
the blank
vrdue is
insignificant.
For the analysis of soil samples having disin-
tegration
rates of the order of 0.01 dis/min/g
or less, more
stringent
conditions
would
have to be observed
in order
to lower
the blank
values
that
are presently
being ob-
served.

Results and Discussion

The soil sample
supplied
to Mound
Laboratory
by
C. W. Sill was
also analyzed
by the
leaching
method
followed
by analysis
of the leached
soil residue
by the
fusion method.
Results of the analysis of two 1-g samples
of this soil are given in Table
III.
It
is clear that
the
leaching
failed to remove
all of the plutonium
from the
soil. The percentages
of ‘Wu
recovered
from the spiked
samples by leaching were 17 and 24%, respectively,
while
81 and 78% of the activity was recovered
by fusion of the
23%
recovered
from
the
two
soil
residue.
The
total

samples,
34.6
and
36.1 dis/min/g,
compares
favorably
with the spiked value of 35.4 dis/min/g.
These data seem
to indicate
that
the leaching
method
used here is inade-
~$e
for Plutonium
soil analysis. The preparation
of the
u-spiked
sample,
however,
involved
heating
the soil
for a total of 4 h at 10OO”C after the plutonium
had been
added. Thus it is possible that the plutonium
reacted with
the soil making leaching ineffective.
The data for the fiist of the four soil samples used
in the intercomparison
study
are shown in Table IV. The
plutonium
concentrations
obtained
by
both
methods
compare
quite
favorably.
The 2XPU tracer
recovery
was
slightly higher for the fusion method.
Table
V lists
the data
for the second
soil sample
used in the intercomparison.
Here again the same general
observations
concerning
the -U
tracer recoveries can be
made.
The leach method
gave slightly
higher
“%
con-
centrations,
but the standard
deviations
of the two sets of
data
overlap.
The
results
of the
third
soil sample
are
shown in Table VI. Once again the same general observa-
tions can be made. Here the 2%1
tracer recoveries by the
fusion
method
were
significantly
higher
with
a much
lower standard
deviation
than obtained
in the leach an-
alysis; however, the averages for the plutonium
concentra-
tions show good agreement.
This set of data, as well as the
data
obtained
on the previous
two soil, samples
clearly
show the need for the use of ‘Pu
tracer in these analy-
ses.
Table VII shows the data on the fourth
soil sample.
Here
the
average
~
concentrations
do not
show as
good
agreement
as the previous
samples
although
from
the
spread
in the
individual
determinations,
especially
with
the fusion
results,
it cannot
be concluded
that
the
results
disagree.
It should
be noted
in Table
VII
that
aliquots
of 50,20,
and 10 g were analyzed
and that as the
aliquot
size decreased
the standard
deviation
increased.
The overall average value for the 11 leached
samples was
13.9 * 4.7 dis/rnin/g.
The average value for 50-g aliquots

66

TABLE 111

ANALYSIS OF A ‘%% SPIKED SOIL SAMPLEa
BY LEACHING AND FUSION

Aliquot
‘h
Removed
‘%% Found
Sample
Total
Analyzed
by Leaching
in Leach Residue
Recovered
Number
(8)
(dis/min/g)
(dis/min/g)
(dis/min/g)

1
1
28.7
34.6
(:;2)
(81%)

2
1
27.5
36.1
(;4;)
(78%)
.—
———
—...

%piked
soil sample (35.4 dislminlg) suppUedto Mound Laboratory by C. W. Sill, Health ServicesLaboratory,
ldaho Fd]s, Idaho.

TABLE IV

23%
DISINTEGRATION
WTES
IN SOIL SAMPLE NO. 1
BY LEACHING
AND FUSION

238Puby Fusion (l O-galiquot)
‘*Pu by Leaching (So-g aliquot)

(dis/min/g)
(%%%
recovery)
“(dis/min/g)
(% ~Pu
recovery)

0.040
16
0.027a
0.051
58
62
o.039a
0.034
71
58
0.037
0.038’
56
43
0.038’
0.036
42
46
o.031a
71
0.050
94
0.045
12

Ave.
0.039
50
0.040
57

Std.
* 0.007
k 27
* 0.008
* 12
Dev.
.——
——
———
——.-—-.
—

aBased on the analysls of an aliquot
of a leach solution
from a 1000-gsample.

.

67

TABLE V

%%
DISINTEGRATION
RATES IN SOIL SAMPLE NO. 2
BY LEACHING AND FUS1ON

238Puby Leaching (50-g aliquot)

(dis/min/g)
(% ‘Ml% recovery)

0.104
42
0.255
25
0.219
11
0.194
77
0.266
30
0.148
111

Ave.
0.198
49

Std.
Dev.
* 0.063
*
38

‘al% by Fusion (10-g aliquot )

(dis/min/g)
(% 2MPu recove~)

0.203
54
0.164
58
0.144
71
0.186
70

0.174
63

TABLE VI

“%
DISINTEGRATION
RATES IN SOIL SAMPLE NO. 3
BY LEACHING AND FUSION

238Puby Leaching (50-g aliquot)

(dis/min/g)
(% ‘MI% recovery)

1.66
22
1.81
13
1.46
23
1.97
100
1.30
63
1.58
89

Ave.
1.63
52

Std.
Dev.
* 0.24
*
38

238Puby Fusion (1O-galiquot)

(dis/min/g)
(% 23’% recovery)

1.75
81
2.06
82
1.41
81
1.34
73
1.38
77

1.59
79

* 0.31
*4

.

68

TABLE VII

23%1 DISINTEGRATION
RATES IN SOIL SAMPLE NO. 4
BY LEACHING AND FUSION

28Pu by Leaching

(% 2~Pu
(dis/min/g)
(aliquot,
g)
recovery)

16.41
15.90
14.56
10.05
11.05
24.31
9.52
14.09
9.92
8.97
18.09

50
50
50
20
20
20
20
20
20
20
20

48
25
40
41
24
40
31
36
51
41
49

Ave.
13.90
39

Std.
Dev. * 4.7
*9

was
15.62
k 0.96
dis/min/g,
for
20-g
aliquots
13.3
t
5.4
dis/min/g,
and
for
10-g
aliquots
26.0 * 22.2 dis/min/g.
A summary
of the ‘%
disintegration
rates for the
four soil samples analyzed
is given in Table VIII. There is
good agreement
between
the average 2?u
disintegration
rates for the first three samples indicating
good agreement
between
fusion and leaching.
Even with sample number
4
where the averages are 26.0 and 13.9 dis/min/g,
consider-
ing the
large standard
deviation
as stated
previously,
it
cannot
be concluded
that
the results
do not
agree. The

Sample
Number

1
2
3
4

‘8Pu by Fusion

(%%%

(dis/min/g)
(aliquot,
g)
recovery)

11.49
16.58
11.28
11.67
65.2
25.35
9.85
56.4

10
10
10
10
10
10
10
10

86
84
87
93
89
75
76
63

26.0
82

* 22.2
* 10

larger
fusion
value could
well have been
caused
by the
fact
that
two
of the
samples
taken
for fusion
analysis
contained
a relatively
large individual
particle
of pluto-
%’u02
particle
1.35 Vm in diam
nium dioxide.
A single
would add about 500 dis/min to a soil sample. This would
increase the concentration
of activity
in a 10-g sample by
50 dis/rnin/g,
while the effect
on a 50-g sample would be
only
10 dis/min/g.
Thus,
it is possible
that the two sam-

23Spu
concentration
contained
a rela”
pies giving a high
tively
large plutonium
dioxide
particle
while the other
sample did not. It should be noted
that the average fusion

TABLE VIII

SUMMARY 23SPUDISINTEGRATION
RATES
BY LEACHING
AND FUSION

Leaching Procedure
Fusion Procedure

Rel.
Rel.
23s~
Std.
DeV.
No. of
23%
Std. Dev.
No. of
(dis/min/g)
(%)
w!!lw
(dis/min/g)
(%)
Samples
—.

0.039
18
8
0.040
20
4
0.198
34
6
0.174
1“5
4
1.63
15
6
1.59
19
5
13.9
34
11
26.0
85
8

69

I

value is 14.4 dis/min/g when these two high values are not
used
in calculating
the
average.
This
average
compares
quite favorably
with the leaching value of 13.9 dis/rnin/g.
This particle
size problem
is more severe when an-
alyzing
for
ZWPu as compared
tO ‘%
because
of the
considerable
difference
in specific activity
between
these
two
isotopes.
Plutonium-238
has a specific
activity
of
3.81
X
107
dis/min/~g
compared
to
1.36 x 10s dis/min/#g
for ‘%.
It should also be noted
that
in
all of
the
fusion
results
the
relative
standard
deviation
is greater
than the standard
deviation
that was
obtained
when
the
spiked
soil
sample
was anlayzed
(6.3%).
This indicates
a sampling error which could also
be explained
by the existence
of small 2~u
particles
in
the soil.
As a further
study
on a possible difference
between
the leaching and fusion procedures
in determining
-u
in
soil,
residues
from
eight
20g
samples
of
soil
sample
number
4, analyzed
by the leaching procedure,
were an-
aly~d
by the fusion
procedure.
The results are given in
Table
IX. With
this
soil sample,
it is seen that
on the
average approximately
93’% of the ‘%
is leached
from
the soil. Also the fact that these eight analyses showed an
average tracer recovery of 39%, not including the leaching
operations,
suggests that
the major
losses in the original
leach procedure
were not in the leach step but, rather, in
the chemistry
that follows.
In conclusion
it appears that the leaching and fusion
methods
in the
present
study
for the determination
of
%%
in soil
agree.
However,
additional
data
will
be

accumulated
in order
to evaluate
this assumption.
Future
plans include
the analysis of additional
leached
soil sam-
ples by the fusion procedure
to determine
whether
or not
leaching
has
failed
to
remove
significant
amounts
of
plutonium
from the original soil sample.

Acknowledgments

The authors
would
like to acknowledge
the assist-
ance of the personnel
at Mound Laboratory
who contrib-
uted
to the determination
of the data presented
in the
report:
M. L. Curtis,
R. Brown,
K. E. DeVilbiss,
J. A.
Doty,
L. C. Hopkins,
V. C. Lacy,
L. G. Musen,
E. B.
Nunn,
R. L. Ryan, and F. K. Tomlinson.
We would also
like to acknowledge
John
H. Harley,
Director,
the New
York Operations
Health and Safety
Laboratory
of the U.
S. Atomic
Energy Commission
for supplying Mound Lab-
Zsspu tracer ~d
for helpful
discussions
orato~
with
the
on the leaching method.
Finally, we also acknowledge
Dr.
C. W. Sill and the Health
Services and Safety Laboratory
of the U. S. Atomic
Energy Commission
at Idaho
Falls,
Idaho,
for
supplying
us with information
on the deter-
mination
of plutonium
in soil by a fusion method.

References

1. J. H. Harley (Ed.), Health and Safety Laboratory Manual of
Stan&rd
Procedures,
U. S. Atomic
Energy Commission
NY04700. Rev. 1970.

TABLE IX

‘%
DISINTEGRATION
RATES IN
LEACHED RESIDUES
OF SOIL SAMPLE NO. 4

Sample
Number

44
4-5
4-6
4-7
4-8
4-9
4-1o
4-11

238&

Leacheda
(dis/min/g)

10.05
11.05
24.31
9.52
14.09
9.92
8.97
18.09

Z8~
in

Leached Soil
by Fusionb
(dis/min/g)

1.10
1.01
0.74
1.13
1.16
0.68
0.62
1.18

Total
‘8Pu in
Sample
(dis/min/g)

11.15
12.06
25.05
10.65
15.25
10.60
9.59
19.27

Average
13.25
0.95
14.20

Std. Dev.
*
5.41
* 0.23
*
5.41
.—.
—
.—--—.
—-.
—--.——

238pu

Recovered
by Leaching
(%)

90.1
91.6
97.1
89.4
92.4
93.6
93.5
93.9

92.7

?
2.4

.

a ‘Pu
tracer recoveries
averaged 39 * 9% for leaching.

b 236Pu tracer recoveries
averaged 76 k 5% for fusion.

70

.

2.

3.

4.

5.

F. E. Butler, “Determination of Uranium and Americium-
Cerium in Urine by Liquid Ion Exchange;’ AnaL Chem., Vol.
37, pp. 340-342 (1965).

L. C. Henley, “Urinalysis by Ion Exchange;’ presented at the
Eleventh Annual Bio-Assayand Analytical chemistry Confer-
ence, Albuquerque, New Mexico,October 7-8, 1965.

1. A. Dupzyk and M. W. Biggs, “Urinalysis for Curium by
Electrodepoaition;’ presented at the Sixth Annual Meeting on
Bio-Assay and Analytical Chemistry, Santa Fe, New Mexico,
October 13-14, 1960.

N. Y. Chu, “Plutonium Determination in Soil by Leachingand
Ion Exchange Separation;’ AnaL Chem., Vol. 43, pp. 449452
(1971).

6. C. W. Sill and K. W.Puphal, “The Determination of Plutonium
in soil:
to be published, (Health Services Laboratory, U. S.
Atomic Energy Comrrdssion,Idaho Falls, Idaho).

7. C. W. Sill arrd R. L. Williams,“Radiochemieal Determinations
of Uranium and the Transuranium Elements in Prwess Solu-
tions and Environmental Samples;’ AnaL Chem., 41, 1624
(1969).

8. C. W. Sill. “Serraration and Radiochemical Determination of

9.

Uranium and &e Transuranium Elements Using Barium Sul-
fate: Health Physics, 17,89 (1969).

K. W. Puphal and D. R. Olsen, “Electrodepsition
of Alpha-
Emitting Nuclides from a Mixed Oxalate~loride
Electrolyte/’
to be published, (Health Services Laboratory, U. S. Atomic
Energy Commission, Idaho Falls, Idaho).

71

.

Introduction

COMMONALITY
IN WATER,
SOIL,
AIR,
VEGETATION,
AND
BIOLOGICAL

SAMPLE
ANALYSIS
FOR
PLUTONIUM

by

Robert A. Wessman, W. J. Major, Kim D. Lee, and L. Leventhal

TRAPELO/WEST
Division of LFE Corporation

Richmond,
California

ABSTRACT

Plutonium
analyws have been performed
at Trapalo/West
for over twenty

years. In recent times, procedural changes have been made to obtain common-

ality in methods for analyzing Pu in different matrices. Procedures used for Pu

environmental
samples such as water, soil, air, vegetation, and biological and

marine samples are discussed. Initial steps involve total dissolution, Ieachhg or

226pu. Tra@r
is used
in all casessince it
ashing, and equilibration
with tracer

results in the most reliable data. An anion exchange procedure is the basic part

of the purification,
An efficient electrodeposition
step permits plating in ten

minutes. Radioactivity
measurements are made using either Frisch Grid ioniza-

tion Chambers or surface-barrier detectors.

Specific
problems
likely
to
be
encountered
in
plutonium
analysis are

discussed. Problems encountered
in measuring and stating error limits at very

low levelsso that they may be used practically are discussed.

... ....... ... .. .. .. . ....... ..... ............ ........ ...

Plutonium
radiochernical
analyses
have
been
per-
formed
at
Trapelo/West
for
over
twenty
years.
Major
changes
are
due
to increased
knowledge
of the
tracer
chemistry
of
plutonium
as well
as the
availability
of
efficient
separation
chernic~s
and reagents
and improvem-
ents
in
nuclear
measurements.
Improvements
have
usually been gradual and metamorphic
rather
than sensa-
tional. The net effect
has still been dramatic.
At one time
our laboratory
had its own calibrated
radiocherru”st~
Use
of tracer
and low- level, high-resolution
alpha spectrom-
etry
have
permitted
the
greatest
improvements.
The
present
state of the art permits practical measurements
to
a counting
error of t 5% at levels as low as 1?4 dpm for a
1000-rnin
count.
That can be reduced
to ?4 dpm if three-
days detector
time per sample is available, etc.

Analytical System

Trapelo
feels that the entire
analytical
system used
must be considered
as a whole. This is even more impor-
tant
in radiochemistry
than
in routine
analytical
chem-
istry.
Chemistry
procedures,
though
most often
stressed,
are only a portion
of the total.
In a small laboratory,
the system might consist of
only one worker
and related
equipment
and procedures.
At another
facility,
such as Trapelo,
the responsibilities
might
be spread
out
according
to the expertise
of each
person.

The
Trapelo
IAmratory
System
for
Plutonium.
What
is considered,
at Trapelo,
to
be the
key
to
the
analysis
of the actinides,
particularly
plutonium,
is listed
(Fig. 1) and outlined
in further
detail below.

73

1.

2.

3.

4.

5.

6.

7.

8.

1.
2.
3.
4.
5.
6.
7.
8.
9.

FIGURE
1

KEY
REQUIREMENTS
FOR
PLUTONIUM
ANALYSIS

Personnel
Low Level Lab and Equipment
Solubilization
(or Leach) of Pu
Accurately
Standardized
23’% Tracer
Equilibration
Decontamination
and Purification
Alpha Spectrometer
System
Standard
Data Calculation
Quality
Control

PERSONNEL
a.
Experienced
in use of procedures
b. Felxibility
in doing different
analyses

LOW LEVEL LABORATORY
AND EQUIPMENT
a. Low Level control
b. Good housekeeping

SAMPLE SOLUBILIZATION
(OR LEACH)
a. Specific procedures
for different
matrices

ACCURATELY
STANDARDIZED
PLUTONIUM-236
TRACER
a. Against an absolute
basis
b. Precision of * 1.5%
c. %
impurity<
0.5 alpha %
d. ‘~
impurity<
0.09 alpha%
e.
Impurity
content
known for correction
purposes

EQUILIBIL4TION
a.
Exchange with tracer during solubilization
or subse-
quently

DECONTAMINATION
AND PURIFICATION
a. The minimum
chemistry
to obtain
weightless elec-
trodeposit
b. Chemistry
tested to remove other actinides
c. Obtain radiochemical
yields of 40 to 90%

ALPHA SPECTROMETER
SYSTEA4
a.
Frisch Grid or surface barrier
b. Resolution
20 to 40 keV
c. Efficiency
30 to 48%
d. No tailing of peaks at baseline

STANDARD
DATA CALCULATION

9. QUALITY CONTROL
AND EVALUATION
a.
Routine
blanks and standards
b. Alpha
spectrometer
checks
on
background
effi-
ciency, etc.

Basic Procedures

The basic procedures
used at Trapelo/West
for low-
level plutonium
are, in many aspects, similar to those used
at many
other
laboratories.
The analyst
has a wealth
of
proven
analysis
sequences
to chose from in assembling
a
set for routine
use in his own laboratory.
Figure
2 is a
schematic
showing how different
sample matrices
fit into
the processing.

Sample Preparation.
The preparation
of samples for
analysis
at Trapelo
follows
generally
accepted
practice
using drying, ashing, grinding,
etc. At Trapelo,
the speci-
mens may be received at the analysis laboratory
in various
states
of preparation,
ranging from a raw sample
to an
ashed residue.

Sample Solubilization.
Solubilization
of the sample
is a critical
part
of plutonium
analyses
using tracer.
In-
deed,
much
time
is spent
in
achieving
this. Within
a
sample
category,
maverick
samples
are
always
found
which will not completely
dissolve by routine
treatment.
The
radiochemist
treats
these
individually
to
dissolve
residuals.
Usually HN03 -HC1-HC104-HF or fusion is used.
In the special case of soil leaching, the procedure
of
Norton
Chu of HASL1
is used and of course
complete
solution
of the soil is not expected.

WAltR
AIR

II

son
SOIL

I

FOOD

II

Bow

FILTIR
NEACIU
IDISSCVAI
VEG.
SWIL
II
B IOL I
i
+
i
1
PRIPARAIION
I SOLUBILIZAIION
/ rWILIBRAIION
1
I
I
I
I
8
wapn.
WI
Gsh
HMJ’;.HCI
tiF
dr$.g
Ming
&ylng
with
and HF
leach
Ht+3,.HCl
Ghlnq
WI
am
WI
ash
Htq
1*#

cycle
cyies
WEI asii
md
HF
1

E

CUAR
S4LUTION

1S1 ANION
CULUMN

2N0 ANION
COLUMN

PURIFIEO
SOLUTION

ELECIROOEFWSITION

ALPIIA
SPEC

INFERP.
ANO CALC

OMPlflED
DATA

Fig. 2

.

.
a. Consistent
interpretation
of spectra
b. Realistic assessment
of precision

74

.

Chemical Rocerhmm.
We wish to attain
a unified
procedure
for
environmental-type
samples.
After
the
samples
have been
solubilized,
it is possible
to use the
same purification
steps for the remainder
of the analyses.
The steps used are not severely influenced
by the original
sample matrice
or the amount
of sample. This commonal-
ity
of
methods
minimizes
having
to
cope
with
many
different
procedures.
Also,
less special
equipment
and
special work areas are required,
different
sample types can
be processed
simultaneously,
and less training
and break
in of laboratory
personnel
is required.
The salient features of the chemical procedures
used
at Trapelo
are outlined
in Fig. 3. The unifying
steps are
anion exchange,
evaporations,
and boiling. The anion ex-
change
procedure
is very similar to that used in the soil
leach procedure
of Chu, who credits
his scheme
to one
suggested by the work of Kressin and WaterbuW.2
The ability to use ion exchange and exclude precipi-
tations,
especially bulky alkaline precipitations
with phos-
phates, etc., is very desirable.
Features
of the exchange
method
are that the solu-
bilized sample, in a volume of 200 to 1500 ml of approx-
imately
6 N HN03,
is processed
by two sequential
anion
exchange
columns
(Dowex
1-X4) to achieve
a solution
from which plutonium
can be electrodeposited
for alpha
spectroscopy.
The first column
is largest,
its size depends
some-
what upon the volume of dissolved sample. Leachate
from
100 g of soil requires a column 2.5-cm diam by 6 cm. If a
1000-g sample
is leached,
the column
length is increased

Fig. 3

to
12 cm. The actinides
Th(IV),
Pa(V),
U(IV),
NP(M,
and Pu(IV)
are absorbed
by the resin while trivalent
Ac,
Am, and Cm pass through.
The retained
actinides
can be
eluted with 4~ HN03 -0.1~ HF (Chu reports
use of 0.4~
and O.01~ respectively.)
The second,
smaller-sized
anion column (1-cm diam
by
2.5 cm)
is used
for final
clean
up
of the solution.
Again
the
sample,
in
6N HN03,
is loaded
onto
the
column.
The
resin
is then
converted,
successively,
with
6~ HC1 and concentrated
HC1 to the chloride
form. Any
‘I% would elute in the HCI fractions.
An elutrient3
of HC1
containing
w
I is used to reduce and elute Pu(III).
This
provides
plutonium
free of any alpha
emitting
actinides
such as Th, Pa, U, or Np. If there is a very large amount
of
Fe(III)
or other
oxidant,
the
first
column
purification
should be repeated
prior to the HN03 -HCI column.

Electrodeposition.
After
evaporation
and
wet-ash
destruction
of trace
organics,
the plutonium
is electro-
deposited
upon
a
stainless
steel
disc
(220-mm
diam,
250-mm2
plated
area). Platinum
discs are used for highest
accuracy.
The ammonium
chloride
method
described
by
Mitche114 is used. This plating method
has been in use at
Trapelo
for many years and is recommended
as a reliable
procedure
which
is essentially
quantitative
with
only
a
10-min plating time.
Very
clean,
almost
invisible
plated
areas
are ob-
tained
if the purification
is done properly.
The stainless
steel
discs should
not
be flamed
after
plating
since an
oxide coating forms which degrades the alpha spectra.

Alpha Spectrometry.
Samples are counted
on a de-
tector
in either
a battery
of Trapelo
Frisch grid detectors
or a battery
of Ortec 450-mm2
surface-barrier
detectors.
The grids operate
on argon-methane
(P-1 O gas) while the
surface-barrier
detectors
are operated
in a vacuum.
Reso-
lution
of the gridded
detectors
are as low as 20 keV at
5.75 MeV. The same Frisch
grid chambers
in 1963 had
only 45-keV
resolution,
a two-fold
improvement
having
been attained
by modification
of the electronic
compo-
nents. Resolution
of the surface barriers is 50 keV.
Background
in the ‘vu
energy peak varies between
0.004
to 0.018
for the different
detectors.
Background
fluctuations
are due primarily
to statistical
variations
but
can be increased
by counter
contamination
from certain
isotopes.
Melgards
discussed
internal
contamination
of
alpha
spectrometers
due to counting
different
isotopes.
On the Frisch grids, collimators
are used to reduce
base-
line tailing. This also reduces counting
efficiency
from 48
to 35%. Efficiency
on the solid-state
detectors
is 28 to

30%..

Calculations.
Calculation
of
alpha
spectrometry
data
is presently
done
using a combination
of computer
and hand calculations.
A smoothed
alpha spectrum
plot is
produced
by the computer,
incorporating
an energy cali-
bration
line
from
standards
counted
with
the
specific
sample.
The plot
is examined
to determine
the isotopes

75

present
in the sample and the alpha peaks are then inte-
grated
within
preselected
energy regions. Corrections
for
background
as well as apparent
impurities
from the 2%Pu
tracer me made.
Errors
of analysis
are estimated
conservatively
and
all errors are included which could significantly
affect the
users’ confidence
in the
data.
This
treatment
becomes
most
significant
at low (< 1 dpm) activity
levels. Rather
than
use simple counting
statistics,
the error
associated
with correcting
for background,
blank, and tracer
contri-
bution,
is estimated
at somewhat
greater
than that error
indicated
by counting
statistics
alone.
This method
also
assumes
that
some
of the
errors
are not Gaussian
and
there is, therefore,
an increased uncertainty.

Operational Experience

Experience
with this procedure
is discussed relative
to
tracer
yields,
isotope
purity,
and
other
operational
aspects.

Yields. Chemical
yields are generally good. Figure 4
shows yields for several different
biological organs ranging
in weight
from
20 to 600 g. There does not seem to be
any
dependency
upon
weight.
The lower yields
for the
nodes are not believed related to sample type.
Yields for leach analyses of various size aliquots
of
soil are shown
in Fig. 5. Different
soils are included
but
no correlation
of yield with soil type has been made. The
lower
yields
primarily
represent
some
of the fust
soils
analyzed
in a given weight range. Some of the unexpected
difficulties
were usually ironed
out. The yield from
1-kg
soil leaches and 100-g dissolutions
are now expected
to be
in the 70 to 80% range.

Purity
of Plutonium
Plates. Natural
and other
arti-
ficially
produced
alpha
emitters
are
often
present
in

environmental
samples
analyzed
for
plutonium.
If not
removed,
22%
and “Am
will perturb
the 23%
alpha
peak. Thorium-227
would perturb
the ‘Pu
tracer peak.
Uranium-232,
a growth
in ‘Pu
tracer,
is also added
to
each
sample.
There
are other
possible
contaminants
of
minor importance.
An evaluation
of the procedure
for decontamina-
tion
from
four
actinide
elements
was performed.
The
plutonium
fraction
was examined
for impurities
on the
alpha spectrometer.
The
results
are shown
in Fig. 6. The amount
of
impurities
on each plate was close to limits of detection.
An
estimate
of
lower
limits
for
the
decontamination
factors
was made and all were greater
than
or equal
to
2 x 103.
More
exact
factors
could
be
determined
but
larger amounts
of impurity
isotopes must be used.
The decontamination
factors
obtained
indicate
the
procedure
is more
than
adequate
for any expected
en-
vironmental
samples.

Operational
Aspects.
This
scheme
of
analysis
appears
to have the desired flexibility.
The commonality
of methods
is not new but the present
scheme seems to
provide
better
unification
than we have experienced
be-
fore.
As an example,
Trapelo
previously
used a unified
system for processing thousands
of biological,
soils, vege-
tation,
and various
collection
media.
The methods
were
34
me
Chefical
procedure
con-
reported
by W. Major ‘
sisted of a cupferron
extraction,
a hydroxide
precipitation
from
a basic carbonate
media, and another
precipitation
from
NHQOH. An anion
exchange
column
purification,
very similar to the second column used in this report,
was
used
as final
cleanup.
Excellent
results
were
obtained
using those
procedures
but
they
contained
some messy,
intermediate
steps,
i.e. the organics
from the extraction
had to be destroyed
by wet ashing. They were also more
time consuming.

FIGURE
4

TRACER
YIELD
FROM
BIOLOGICAL
ORGANS

Organ

Kidney
Heart
Rib
Node
Spleen
Lung
Liver
Reagent
Blank*

Aliquot
Tracer
Yields
g
Av. of Duplicates

60
87%
65
80%
30
70%
20
38%
560
72%
210
77%
620
98%**
—
91%

.

.

.

*Blank results 0.0 b 0.01 d m “%J.
l?’
*Ylded
beaker cau.wd 2
0 yield on a liter.

76

FIGURE
5

TRACER
YIELD
FROM
SOILS

.

,

Sample
Size
No.
Samples
Yields
Average

g
Range

Leaching

100
10
28-82%
60%
100
5
28-62%
45%
100
5
18-76%
53%
1000
12
8458%
32%
1000
13
3088%
60%
500
6
72-94%
80%

Remarks

Early
work
Later
work*

Dissolution

100
5
42-88%
75%
.-..
———
———
———
——.
—
——. —..

*TWOdifficult soils with 107oyields not included.

FIGURE
“6

DECONTAMINATION
(Tested

Impurity***
Added

FA~ORS
FOR
Pu
on Duplicate
Runs)

Found
on Pu Discs

PROCEDURE

Estimated

kOtODF2
Individual
Av.
Decontam.
dpm
dpm
dpm
Factor

Z3q.h
429
0.06
0.13
23X103
0.20

nlpa
700
0.08
0.06
>
10 x
103
0.05

?’33u
532
0.12
0.30
>2
X103
0.46

Mi~
467
0.06
0.06
>8xlf)3**
0.00

*Results corrected by
Pu tracer yield.
**cm deeontamjnatjon will be sjrnilm to Am. ***Potential interference in PU alpha spec wodd
be:

Pu peak -
Am,
U (minor)
Pu peak -
Th,
Am
Pu peak -
Th,
Cm

77

Specific Problems in Low Level Plutonium Amlysis

A few specific problems
related
to low-level pluto-
nium analysis are given. These are mutual problems
faced
by analysts and which affect the ultimate
data users.

Low-Level
Aspect.
High-level
plutonium
samples
sometimes
appear
unexpectedly
in
analysis
programs.
Sometimes
they have been
prepared
as program
evalua-
tion spikes or other
tests. They are a defiiite
contamina-
tion
hazard
to other
samples
when
this
is not
known
ahead
of time.
in a program
with
mixed
levels, a pre-
monitoring
system must be set up as was done in past soil
analysis programs.c
At this laboratory,
low-level laboratory
operations
suffice
for
analysis
of
samples
ranging
from
zero
to
approximately
100 dpm.
The
greatest
barrier
to
cross
contamination
is the use of new glassware, especially
on
low-level samples.
If the project
work does not merit this
added expense,
then second-hand
glassware from projects
of
similar
or
lower
level can be used.
Used glassware
introduces
another
variable since cleaning procedures
may
not
be
perfect.
Other
sources
of cross
contamination,
such as reagent
bottles,
centrifuges,
platers,
etc. must be
minimized
by good
housekeeping.
Effectiveness
of such
operations
must be monitored
by processing
blanks with
each batch of samples.

Phstonium-236
Tracer. The key to accurate
analyses
at
many
laboratories
is use of 2%Pu tracer.
Stocks
of
tracer
available
have
been
found
to
contain
a slight
‘8Pu
(0.2 to 0.5 alpha %) and
a parent
contamination
of
1’
2 ~u
(0.04 to 0.09 alpha Yo).The contamination
increases
relatively
with
time,
almost
proportional
to the 2.8-yr
‘Pu
decay.
Tracer
purchased
in late
1970
from
the
USAEC,
Oak
Ridge
tests
no better
than
our
previous
stock (produced
in 1963).
The most
serious effect
is in the analysis of ‘~u.
The amount
of correction
needed is difficult
to determine
accurately.
Use of very small amounts
of tracer (3 dpm)
minimizes
the
correction
but longer
counting
times are
required.
An alternate
method,
in a sample with measur-
able ‘%
content,
is to split the sample and analyze one
part,
with
tracer,
to obtain
the
23%
content
and the
second
art, without
tracer,
to obtain
a less unperturbed
&
‘V1.rl”
u ratio.

Evaluation
of
Analytical
Quality.
Radiochemists
may
expound
upon
very good
tracer
yields
and relate
them
to analytical
quality.
Data
users
may
be unduly
influenced
and
give high-yield
data weight
over average
yields.
In low-level plutonium
analysis, a good yield means
that
signal-to-background
ratio
and figure of merit
for a
given sample is being maximized.
This is important,
but a
very high yield, say 96%, may be an artifact,
particularly
in diode counting.
It should not outweigh
a yield of 80%
or even 50%.

Melgard
discussed
the
factors
affecting
the
effi-
ciency
of both
Frisch
grid and surface-barrier
detectors.
Non-uniformity
of plating
can result in a 20’%variation
in
counting
efficiency
on s.b. detectors.
Sample positioning
has a large effect on efficiency
at short sample-to-detector
distances
used on surface barriers.
The error is greater for
smaller
detectors.
Frisch-grid
efficiency
is insensitive
to
these variations.
Thus yields on surface-barrier
detectors
may not be absolute,
but
since yield cancels
in isotope
dilution
analysis,
this is not important.
At Trapelo,
tracer
yields are considered
accurate
to * 3% on Frisch grids and
A 5% on
the
450-Inrrt2
surface
barriers.
Thus,
a sample
with 96% yield on solid state could measure 88% yield on
a Frisch grid, with no harm, except
to the self esteem of
the radiochemist.

Summary

A commonality
in methods
of low level plutonium
analysis
in environmental
samples
at Trapelo/West
has
been
briefly
described.
Emphasis
has been
placed
upon
aspects
considered
most
important
to obtain
accurate
results.
The
system
of analysis
at a given laboratory
is
considered
to be most important.
Providing
certain
primary
operations
are
accom-
plished
in
an
analysis,
intermediate
processing
steps
assume
secondary
importance,
provided
tracer yields are
reasonable.
Most important
is use of an absolutely
standardized
plutonium
tracer
and
equilibration
in
the
sample.
As
sample activity
levels decrease,
spectra
interpretation
and
data calculation
methods
assume greater importance.
The
data user should use low level data with large error limits
with
caution.
In the haste of project
analysis, data users
rather tend to disregard error limits.
Large error
limits,
calculated
by routine
statistical
methods,
should
be verified
in empirical
tests
such
as
dilution experiments,
blanks, etc.
With these
considerations,
it is recommended
that
promulgation
of approved
methods
by any agency group
or
project
be
done
with
caution.
That
flexibility
of
methods
be allowed
to each laboratory
system and that
emphasis
for correctness
be placed
upon
obtaining
the
same results on the same material by independent
labora-
tory
systems.
This is already
the basis of operation
for
some of the most successful data-gathering
systems in the
nation.

References

1.

2.

Norton Y. Chu, “Plutonium Determination in Soil by Leaching
and Ion-Exchange Separation,”
Anal. Chem. 43, 449-542
(1971).

L K. Kressin and G. R. Waterbury, “The Quantitative Separa-
tion of Pu from Various Ions by Anion Exchange;’ Anal.
Chem. 34,1598-1601 (1962).

.

.

.

.

78

.

.

3. W. J. Major, R. A. Wessman, R. Melgard and L. Leventhal,
5. R. Melgard, R. A. Wessman and L. Leventhal, “The Relative
“Routine Determination of Pu by Tracer Techniques in Large
Advantages of Gas and Diode Detectors in Low Level Alpha
BiologicalSamples;’ Health Phys. 10,957965
(1964).
Spectroscopy;’
14th Annual Bioassay and Analytical Chem-
istry Meeting,New York City, Oct. 7-8, 1968.
4. R. F. Mitchell, “Electrodeposition of Actinide Elements at
Tracer Concentrations,” Anal. Chem. 32, 326-328 (1960).
6. W. Major, R. Wessman,R. Melgardand L. Leventhal, “Routine “
Determination of 239PU in Fu~~
soil Lattices by Tracer Tech-
niques:
Tenth Annual Meeting Health Physics Society, Los
Anageles,Calif., June 1965.

.

79

.

“

.

THE PARTICLE
PROBLEM
AS RELATED
TO SAMPLE
INHOMOGENEITY

by

Claude W. Sill

Health Service Laboratory

U. S. Atomic
Energy Commission

Idaho Falls, Idaho

ABSTRACT

The effect of the specific activity of single particles of various sizes on

the
comparative
homogeneity
of
plutonium
distribution
in soil samples is

discussed.

Information
is presented on the relative efficacy of leaching procedures
versustotal sample decomposition as a function
of particle size and origin.

. ............. .. . . .. .. . .. . .. ... .... ................

The activity of N spherical particles of pure ‘i~uOz
is 0.721 N D3 dpm where D is the diameter
of the particle
in microns.
Because
the
activity
is proportional
to the
third
power of the diameter,
a ten-fold
increase in diam-
eter
gives a thousand-fold
increase
in
activity.
If the
activity
is low, as is presently
true with average soils, the
entire activity
could have resulted
from a very few parti-
cles of
reasonable
size,
making
reproducible
sampling
virtually
impossible.
For example,
a single 1# particle
in
10 g of
SOil gives an average
activity
of 0.072 dp~/g.
Levels around
0.04 dpm/g are widely encountered
in the
environment,
while levels as high as 1 dpm/g have caused
considerable
concern
among
some
critics.
These
levels
could have resulted
from single particles having diameters
of 0.82 and 2.4 g, respectively,
in 10 g of soil. A single
large
particle
would
contribute
as much
activity
as a
thousand
smaller ones with one-tenth
the diameter.
Con-
sequently,
different
solid
aliquots
of the
same
sample
submitted
for
analysis
could
give results
differing
by
many
orders
of magnitude
depending
on the number
of
particles
present
and
their
size
distribution
in
each
aliquot.
Larger
samples
would
obviously
help
obtain
a
more
representative
mean
but
would
not
eliminate
the
problem.
In laboratory
measurements
of the characteristics
of
aerosols
resulting
from small-scale
burning
of plutonium
metal
and
alloys,
Ettinger
et all
found
mass median
diameters
(mmd) of 0.03 to 0.14 ~. They also quote work
of others giving mmd’s of several v for other
conditions.
Mishirna and Schwendiman2
found
a mmd
of 4.2p
for
aerosols
from ignition
of large metal ingots in moderate

airflows,
and mmd’s up to 60 w for the airborne
material
resulting
from heating dry plutonium
compounds
in flow-
ing air streams.
Similarly,
Kelkar and Joshi3 found pluto-
nium
particles
with
a median
diameter
of
1.1 p in a
laboratory
handling
plutonium
compounds.
It
seems
entirely
reasonable
to
expect
severe
sample
inhomo-
geneity
at short
distances
from plutonium
facilities,
par-
ticularly
if the activity
levels are high, and a detectable
problem
even at considerable
distances.
Fowler
et al.4
show results varying from O to 778 dprn/g in a single soil
sample collected
near the impact area of an aircraft
carry-
ing a nuclear device.
On the other hand,
if the particles are even as small
as 0.1 p at least 556 particles
of the pure oxide would be
required
in a 10-g sample to produce
an average level of
even 0.04 dpm/g.
Such a large number
of small particles
should
permit
the
sample
submitted
for analysis
to be
homogenized
and sampled
better
than the statistical
un-
certainty
associated
with either the subsequent
analysis or
the environmental
sampling
itself. However,
most of the
globally distributed
plutonium
results
from detonation
of
nuclear devices that give particles only a few r-rwin diam,5
an extremely
large number
of which would be required
to
account
for the observed
activity.
Furthermore,
material
from
the
detonation
of nuclear
devices
will have been
completely
vaporized
and
recondensed
giving particles
containing
a very small fraction
of plutonium
rather than
separate,
discrete
particles
of
the
pure
oxide.
Conse-
quently,
little
inhomogeneity
of consequence
might
be
expected
on soils containing
only plutonium
from global
fallout, even on only 10-g samples.

81

Although
few
in
number,
the
experimental
data
shown in Table I appear to substantiate
the correctness
of
the above reasoning.
The first two samples were obtained
near a plutonium
facility,
but one which was not known
to
have
released
any
plutonium
to
the
environment.
Samples 3 through
7 were obtained
at distances of about
2.0,
2.0,
16,
17 and
43
miles, respectively,
downwind
from a plutonium
processing
facility
known
to have re-
leased a significant
quantity
of plutonium.
Samples 8 and
9
were
taken
at
distances
of
about
50 miles
and
100 yards,
respectively,
from
two other
facilities known
to
have
released
plutonium.
Every
result
obtained
on
samples 1,2, and 7, and all but one result each on samples
5, 6, and 8 are well within
the statistical
uncertain
y of
the
analyses
on
10-g samples.
The
plutonium
present
probably
resulted
entirely
from global fallout.
However,
the single, high values in samples 5, 6, and 8 are 1.5 to 4
times
the
other
values in the
same sample
and clearly
represent
a
significant
difference
in
that
particular
aliquot,
possibly
caused
by a single, larger particle.
The
results
on samples 3 and 9 show the pronounced
hetero-
geneity
to be expected
on samples taken relatively
close
to the source
where
larger particles
might
be expected.
Samples
3 and
4 were
taken
at greater
distances
than
sample 9 but the source was much larger and the area is
subject to fairly high winds.
The
particle
problem
becomes
particularly
acute
with
_u02
for which
the
numerical
constant
in the
above activity-particle
size relationship
is 202. In a 10-g
sample,
single particles
of 0.1- and 1-g diam give average
activities
of 0.02
and
20.2 dpm/g,
respectively.
Conse-
quently,
even low-activity
samples might be expected
to
give extremely
erratic
results occasionally
due to sample
inhomogeneit y,
particularly
in the
vicinity
of facilities
handling
Zspu
where
larger
puticles
might be encounter-

ed.
In
one
such
example,
the
ratio
of ‘%
to
‘%
changed
from
1.6 to 0.15 on two separate aliquots
of the
same sample showing conclusively
the presence of discrete
particles of different
composition.
The numerical
constant
in the activity+ize
expres-
sion above is only 6.94 x 10-4 for a highly enriched
UOZ
containing
170‘U02
and 99$Z0‘SU02.
A single particle
of
10W diam would
produce
activity
in 10 g of soil of
only
0.069 dpm/g.
Consequently,
relatively
larger parti-
cles are required
to produce
significant
activity
in a few
particles
and the particle
problem
is expected
to be rela-
tively
small
for
uranium
oxide,
even when
highly
en-
riched.
The
constants
are
8.03 x 10_3 for “Np02
and
41 1 for MlAm02,
giving
rise to particle
problems
inter-

mediate to those described above.

References

1. H. J. Ettinger, W.D. Moss,and H. Busey,Nucl. Sci. Eng., 30, 1
(1967).

TABLE
I

REPRODUCIBILITY
OF ANALYSES
USING
1O-GRAM ALIQUOTS
OF
PREPARED
SOILS

Number

1

8

9

aLeached
according
acid soluble, 90%.

Pu found,
(dpm/g)

0.110
* 0.009
0.116
* 0.010
0.112
f
0.012
0.101
* 0.008
0.111
* 0.008

0.060
* 0.007
0.050
f
0.007
0.054
* 0,008
0.063
+ 0.007

1.59
* 0.04
0.56
t
0.02
0.94
* 0.03
0.68
* o.03a

0.62
? 0.02
0.56
+ 0.02
0.57
f
0.02

0.044
* 0.006
0.077
+ 0.008
0.042
+ 0.005
0.055
* 0.010
0.047
i
0.006

0.079
k 0.009
0.058
+ 0.008
0.071
* 0.009
0.29
+ 0.01

0.051
* 0.007
0.066
* 0.009
0.056
* 0.006
0.052
* 0.006

0.071
f
0.008
0.22
k 0.02
0.051
* 0.007
0.059
* 0.006

0.35
* 0.02
0.78
* 0.04
1.73
* 0.04
0.26
* 0.01

to Ch U3
. Insoluble, 10%;

82

2. J. Mishimaand L.C. Schwendirnan,“The Amount and Charac-
4. E. B. Fowler, J. R. Buchholz, C. W.Christenson, W.H. Adams,
teristics of Plutonium Made Airborne under Thermal Stress,”
E. R. Rodriguez, J. M. Celrna,E. Iranzo, and C. A. Ramis, U. S.
Symposium on Health Physics Aspects of Nuclear Facility
Atomic Energy Comm., Document LA-DC-9544(1968).
Siting, Idaho Falls, Idaho, Nov. 34,1970.
5. A. W. Klement, Jr., Ed., “Radioactive Fallout from Nuclear
3. D. N. Kelkarand P. V. Joshi, Health Phys., 19,529 (1970).
Weapons Tests:’ U. S. Atomic Energy Comm. Symposium
Series5,98-143 (Nov. 1965).

83

PLUTONIUM
IN SURFACE
SOIL
IN THE
HANFORD
PLANT
ENVIRONS

by

J. P. Corley, D. M. Robertson and F. P. Brauer

Battelie Memorial Institute

Pacific Northwest Laboratory

Richland, Washington

A8STRACT

Surface
soil sampling from
February,
1970
through
April,
1971
on and
around
the
Atomic
Energy
Commission’s
Hanford
Reservation is described.

The sample sites selected were from
lessthan 1 mile to asfar as 30 miles from
major plutonium-handling
facilities, including sitesaround the perimeter of the

AEC controlled land.

The top one-half
inch of soil was sampled. Vegetative litter and rootmat
were avoided as much as possible. Portions of the mixed
soil samples were

dried and analyzed for
plutonium
content, using acid leaching, solvent extrac-

tion, and alpha counting. Several locations were sampled in replicate. Certain

samples were analyzed
in duplicate. The plutonium
results (all as dpm pluto-

nium
per g of dry
soil) grouped
by general location were: within
restricted

areas, from
0.05
to
1.4; outside restricted areas but within
the reservation

boundaries, <0.01
to 0.28; and outside the plant boundary,
from
<0.01
to

0.13.

Introduction

I must preface my talk with a cautionary
remark.
Although
any conclusions
that might be drawn from the
limited
data
we
have
available
so
far
can be at best
tentative,
we believe
that
recent
data
collected
at the
Hanford
site, and the techniques
used, might be of inter-
est to this symposium.
Analyses for plutonium
in air, water, and foodstuffs
have
been
part
of the
routine
surveillance
program
at
Hanford.
We have surveyed the ground and other surfaces
for plutonium
where
there was possible deposition
from
stack emissions,
waste spills, etc., using direct instrument
measurements.
Detectable
plutonium
deposition
from the
few such incidents
has been confined
to restricted
areas.
The surface contamination
level that can be detected
with
our
portable
instruments
is
approximately
0.007 WgPu/100
cm2.
The desire to obtain
additional
information
regard-
ing any spread of plutonium
beyond
the restricted
areas,
as well as to distinguish
between
any plutonium
in soil
resulting
from
plant
activities,
and
that
resulting
from

fallout
led to a screening
survey for plutonium
in surface
soils both on- and off-site in February,
1970.
The results to date and the procedures
followed are
discussed in this paper. Although
some additional
samples
have been taken,
the major part of my discussion
will be
on the initial
survey.
The limited
amount
of subsequent
data has given results within the same range of plutonium
concentrations.

Sampling Procedure

h
order to minimize
the variables associated
with
the
sampling,
an attempt
was made
to
select
uniform
sampling sites. At Hanford,
this means desert soils as free
as possible
from rocks and standing
vegetation.
Emphasis
was placed on the sampling
of undisturbed
soils and only
minimum
amounts
of rootmat
or vegetative
litter
were
accepted.
Since the primary
objective
was to determine
the current
distribution
of plutonium
rather than to make
a total
environmental
inventory,
the sampling
depth
was
kept to a practical minimum.

85

Initial
sampling
was
done
with
a flat-bottomed
scoop
approximately
18 by 12 in. An attempt
was made
to take only the top !4 in. of soil. Subsequent
sampling
has
been
done
with
a closed-top
sampler
to
minimize
variation
in sample
depth.
An ordinary
cellulose
tape
container
gives a neat,
sharp-edged,
reproducible
cut in
our desert
soils, 9-cm in diam by 1.6-cm deep, provided
no larger gravels are present. A trowel was used to make a
clean cut across the bottom
edge of the container
and to
retain the entire sample for transfer
to a tared polystyrene
sample
jar.
Repeated
cuts
within
the selected
sampling
area give a total sample weight of 150 g or more. Samples
of known
depth
can be taken by removing
soil from the
side
of
the
implanted
container
just
deep
enough
to
expose its bottom
edge, removing the sample, and repeat-
ing the procedure.
Plutonium
analyses
on
the
samples
of February,
1970,
were
performed
by
two
laboratories,
Battelle-
Northwest
and
U.S.
Testing
Company.
The
Battelle-
Northwest
Laboratory
procedure
used
aliquots
of 10 g
(dry weight)
of soil for plutonium
analysis.
Each aliquot
was spiked with a nominal
1 dpm of 2*Pu and heated for
2 to 3 h at 750”C.
The soil was then leached
with both
dilute
and
concentrated
hydrochloric
acid
followed
by
concentrated
nitric acid. The total acid contact
time was
3 to 4 d. The leach solution
(10~
in HC1) was loaded on
Dowex-1 anion exchange
resin and the resin washed with
10~ HC1. The
plutonium
was reduced
and eluted
with
0.1 ~ ammonium
iodide
in
5~ HC1.
The
plutonium-
beanng
effluent
was converted
to 8~ HN03
and again
loaded
onto
Dowcx-1;
the
resin was then
washed
with
8~ HN03,
and eluted with 1.2N HC1. A final purification
was accomplished
with
a thenoyltrifluoracetone
(TTA)
extraction.
The
plutonium-bearing
organic
phase
was
evaporated
on a platinum
disc, counted
with a 150 mmz
silicon
surface-barrier
detector
for
8 to
10 d. Process-
blank
counting
rates
were
less
than
one-tenth
of the
-
lowest
sample
counting
rate.
The detection
level by this
procedure
is estimated
to be 0.01 d m per sample for a
$
10 d count
or about 0.001 dpm/g z
‘~u
of soil. Use of
the silicon
detector
permitted
distinction
of ‘%h
from
the 239+2’%1.
The 137CScontent
of the samples was measured
by
gamma-ray
spectrometry
and used to normalize
the pluto-
nium
results for differences
in the fallout
content
of the
various samples. Several hundred
g of sample were placed
in a 5-in.
diam
by
3/4-in.
deep
plastic
container.
The
samples were counted
for at least 1000 min each between
a pair
of
6-in.
diam
by
5-in.
thick
NaI(Tl)
detectors
operated
in
anticoincidence
with
a
plastic
phosphor
annulus
for Compton
suppression and background
reduc-
tion. A weighted
least-squares
method
was used to calcu-
late 137Csestimates
from the spectral data.1’2’3 The
C~CU-

lations gave a precision
estimate
for the 137CSanalyses of
better than * 5%.
For a few of the February,
1970 samples, and for
all subsequent
samples,
a somewhat
different
plutonium

analytical
procedure
was
used
by
the
U.S.
Testing
Company.*
Samples
were weighed,
oven-dried
at 125°C
for
24 h, and
manually
stirred
to mix
the sample
and
break up any clods. Five g of dried soil were used for hot
leaching.
One-hundred
ml of 8~ HN03
plus 2 drops
of
concentrated
HF were applied under reflux. The resulting
mixture
was
filtered
and
washed
with
hot
1~ HN03.
After
evaporating
the leach solution
to dryness
and re-
solubilizing,
a lanthanum
fluoride
coprecipitation
was
carried
out.
TTA extraction
and electrolyte ic deposition
were used to purify
and mount
the plutonium
for count-
ing. Counting
was performed
by exposing
NTA film to
the plated
disc for approximately
1 wk. Alpha tracks in
the fdm were counted
and converted
to dpm as total Pu.
Yield
by
this
procedure
was nominally
65%, with
an
expected
detection
level of about
0.007 dpm/g
of soil.
The procedure
described
following
the leach step is our
standard
bioassay procedure
for plutonium.

Analytical Results

Figure
1 shows the Hanford
reservation,
the chem-
ical
separations
areas,
the
reactor
areas,
and
the

——-—
______________

*U.S. Test ing Company,
Rich land Branch - A Contractor
to the
Atomic Energy Co remission.

Fig. 1

ik-.

-1
o-i

1.

6’.[{
04

t
.

86

laboratory
areas, as well as nearby communities.
Sampling
sites are indicated.
The
distance
to the
nearest
chemical
separations
area has been listed in Fig. 2 for all samples outside these
areas. Both areas have, in the past, included
facilities
for
liquid
processing
of irradiated
fuels, while the West area
also has facilities
for processing
plutonium
to metal and
metal fabrications.
Much smaller quantities
of plutonium
have been handled in the 300 (Laboratory)
Area.
For the analytical
data presented
in Table I, U.S.
Testing Company
data are identified
with an asterisk; the
remainder are Battelle-Northwest
data.
Bulk density
measurements
of the dried soil ranged
from
1.35 to 1.65 g/ml, and an average value of 1.5 g/ml
has
been
used
to
convert
concentration
by weight
to
surface deposition
per unit area.
The right hand column is labeled Multiple
Analyses.

The entry
aliquot
in this
column
indicates
analysis
of
more
than
one portion
of one sample
and sample indi-
cates analysis of different
samples taken at that one site.
The analytical
results
obtained
on several samples
taken from one sample site generally show about the same
variation
as replicate
analyses
on one
sample.
For
the
analyses
performed
by
the
Battelle
Northwest

Tulf
i

PLLIM#IUf
M SUWACE 6011 Al
HM7021
- KBWJJkf6462fi,
1!2S

(Lsboratov
A except
as ltifcmd
b,
q

mStolce
fro=
A9tlo

Seowat!ms
9“-[239.240]
F.”.:;9..6OI
F“ct{m
P“.grz,;wz)
Are,,
IOleS
:zqq
P..230/Tot*l
?U
nml{catQ

A.
Insld+
Cwdcal
Swarat!ms
At-ras

0.13* 0.01
1.11
0.2s
:0.010

0.0s7
*
0.004
0.49
4.03

0.0s?
2
0.00s
0.49
.0.02

1.23
,
0.06
0.03S
,
0.W3
1.S4
,
0.16
11.9
0.031
,
9.W

0.8s
,
0.04
7.3
O.om
,
0.003

0.21
,
0.01
2.41
0.04;
,
0.036

6.
Utthtn
.anford
Sfte Bo.ndaw

0.030
0.0s
0.026

0.20
0.25

0.11

0.07

0.029
0.026

IZ:Z

0.022
0.026

0 .02s
0.031

0.240
0.019
0.022

0.019

0.023
0.019
0.020

o.m7
%019

0.023

0.023

0.016
0.016

0.01?
0.020
0.024

0 .02s

0.013

Al 4WM6

All wow

56V19S

woks

Al !wots

2
0.27
,
0.01
0.24
,
0.02

G0.10

2
0.033
,
0.033
0.035
*
O.LW
0.03Y
,
O.wz
0.0Z2
,
O.o:z

0.033,
0.010
2.10
0.0s
* 0.007
1.63

.0.09
.0.0s
.0.22
0.26
0.018,
0.39

2
!3.064
, 0.02<

3.051
,
0.0’33
.0.08

0.s03
,0.18

4
0.046
,
0.003
.0.07
0.031
,
0.00s
.0.10
0.03s
$
0.O’3Z
0.37
O.M
,
0.020
-Z. WZ
0.79

a
2.112
,
0.W6
0.96
0.092
,
0.M6

0 .Za9
% 0.018
0.0+6
,
0.012
0.64
O.a
,
0.014

9“
O.ou, 0.001
O.ow, 0.000
0.060,
0.092

9
9.081
,
0.026
J.W5
,
0.005
-3.23s

0.063
,
O.mo
0.76
3.039
,
O.OIZ
1.?6

All wou

11
2.99Z,
O.OM

V.031
.
0.79
0.061
: 0.010
0.Z7

12
0.016
,
0.00Z

16
9.023
,
O.ooz
0.022
,
0.032

c.
off-s!
CR

12
0.061
,
o.ms
17
0.028
,
0.021
0.168
:
0.00?
W.076

20
0,025
-
0.C02

G,0.005

0.14
0.023
: 0.010

4.06
.0.16

0.52
0.026
! 0.0s4
0.76
.
0.027
,
0.006
0.65

0.Z2
0.014
* 0.027
#o. 04

*19S

21
0.024
!
0.003
Q.ozs
0.21
.0.06
0.21

u

x
o

0

0
0

0

Laboratory,
the
results
are generally
within
statistically
expected
range. Values
from
the other
laboratory
com-
pare
less closely.
The
variations
between
sites and
be-
tween
replicate
samples
are
believed
due
largely
to
non-uniform
distribution,
but
other
sources
of inconsis-
tency cannot
be ruled out. However,
the values obtained
are generally
in the expected
range from other
reported
results for plutonium
from fallout.
The primary
intent of the work done was the identi-
fication
of plutonium
from
plant
releases within
the re-
stricted
areas
and
to
determine
if this
plutonium
had
migrated
to areas
outside
the restricted
areas. Figure
2
shows the
239+ -U
activity
concentration
as a function

of
distance
from
plutonium
processing
facilities.
The
variabilityy is apparent,
with no clear relationship.
Figure 3
is a plot
of ~S~~Pu
actitity
concentration
versus that
137cs, attributed
to fallout.
AS is
activity
normalized
to
readily
seen, the samples
marked
W, those
from
the re-
stricted
area containing
the plutonium-handling
facilities,
are distinct
from all other
samples.
These
values have a
Pu/Cs
activity
ratio
which
can be attributed
to plant
releases. The remaining
data, both
on and off the reserva-
tion, have the same nominal
Pu/Cs ration,
a ratio charac-
teristic
of regional surface fallout at the time of sampling.

x

00
0
x

o

ANALYTICAL
LIMIT

x
.— --
--._-----
X
------lr -------

0
5
10
15
20
2s

OISTANCE
IN MILES

F&. 2

87

v
v
v

kv
e

-A
v
PuHANOLING
AREA

%
a
SEPARATION
AREA

s
A
RESERVATION
ANO

~%
A
CONTRCl
PERlhUER

A
O
OFFSITE

-o

“ BNW DATA

1
!
1
I
1
1

“o
Qlo
0.20
fL30

ACTIVITY
RATIO
‘a%240)Pu/137c,

Fig. 3

Summary and Conclusions

During
the operation
of the Hanford
plant, small,
localized
releases
of
plutonium
have
taken
place.
The
samples
from
the restricted
West area, confirm
this fact
and indicate
that a different
Pu/Cs activity
ratio is to be
expected
from that due to fallout. The samples from both
outside
the
restricted
area but within
the plant
bound-
aries,
and
outside
the
plant
boundaries,
have the same
Pu/Cs ratio, indicating
that the plutonium
found is due to
fallout and not plant operation.
I have been impressed
by the precision
reported
in
other
papers at this meeting, and hope to make use in the
future of some of the things we have learned.

References

1.

2.

3.

4.

W.
L. Nicholson,
J. E. Schlosser,
and
F. P. Brmscr, ‘The
Quantitative
Analysis of Sets of Multicomponent
Time - De-
pendent
Spectra
from
the Decay of RadionucIidesY
Afucl.
Instr. andMcth.,2545-66,
1963.

F. P. Brauer and
J. E. Schlosser, “Spectral
Data Handling
Systems,” Modern Trends in Activation Analysis, J. R. DcVoe,
cd., 2, 1102-1107, 1969.

J. F. Fager and J. H. Kaye, “Preliminary
Processing of Multi-
plexed
Two-Parameter
Gamma-spectrometer
Data,”
Deem
Proceedings Spring 1970,45-51,
1970.

Environmental Evaluations Staff, J. P. Corley, Manager, Evah/a-
tion of Radio[ogr”ea[Conditions in the Vicinity of Hanford for
1969. C. B. Wilson and T. H. Essig, eds., November
1970,
BNWL-1505 Appendix.

.

88

MEASUREMENT
OF PLUTONIUM
IN SOIL

AROUND
THE
NEVADA
TEST SITE

by

Wayne Blissand Leslie Dunn

Western Environmental
Research Laboratory

Environmental
Protection Agency

Las Vegas, Nevada

ABSTRACT

Experiments conducted at the Atomic Energy Commission’s Nevada Test

Site between 1951 and 1963, using plutonium
in both critical and sub-critical

configurations,
have resulted in distribution
of plutonium
beyond
the bound-

aries of the Test Site. The Southwestern Radiological Health Laboratory
of the

Environmental
Protection Agency is conducting a survey to assessthe distribu-

tion and concentration of plutonium
in the off-site environment.

Special sampling methods were devised since desert soil is too coarse and

dry for auger and cookie curter sampling techniques, Soil sample analyses are
performed
by
a dissolution,
ion exchange, and electrodeposition
procedure

followed
by alpha spectroscopy. Plutonium
has been detected in four locations

around
the
Nevada Test Site. These
locations correspond
to
fall-out
areas
previously
identified
for
the various test series. Plutonium
concentrations in

the top
3 cm of
soil were 10 to 100 times greater than the concentration
in

soilsfrom areas not subject to contamination
by these series.

Nuclear experiments
conducted
by the U.S. Atomic
Energy
Commission
at
the
Nevada
Test
Site between
1951
and
1963
using
plutonium
in both
critical
and
sub-critical
configurations
have resulted
in distribution
of
plutonium
beyond
the test-site
boundaries.
These experi-
ments
were generaUy
of three
types.
There
were
acci-
dental
ventings
of underground
explosions
which contri-
buted
little,
if
any,
to
off-site
plutonium
deposition.
There
were
also
atmospheric
detonations
of
full-scale
nuclear
explosives,
such as the Plumbbob
series of 1957.
A high percentage
of the plutonium
used in such devices
would
escape unf~sioned.1
These experiments
may not
have contributed
largely to local off-site
deposition.
The
third
type,
and
probably
the
principal
contributor
to
current
plutonium
in the close-in, off-NTS area, were the
so-called one-point
or safety detonations.
These tests were
to
test
the effects
which
would
result
should
the high-
explosive
component
of a device be accidently
detonated.
As part of its responsibility
for radiation
monitoring
around
the
Nevada
Test
Site, the Southwestern
Radio-
logical Health
Laboratory
(SWRHL)
has been conducting

a soil sampling
program
to determine
off-site plutonium
levels. The main
objective
of the study
is to define the
current
plutonium
distribution
around
the Nevada
Test
Site, determine
if it is migrating
by natural
phenomena,
and
determine
if man
has been,
or may be, subject
to
plutonium
exposure.
Should
there be any health
hazard,
it will be shown by the study results. Concurrent
with this
off -site study,
more
detailed
and
complex
studies
are
being conducted
on the Nevada Test Site to evaluate soil
to man routes
and any related
hazards.
Studies
of resus-
pension,
air
sampling,
plant
and
animal
sampling,
and
particle
analysis,
shall be done following
this distribution
survey. Procedures
and results for the early phase of the
off-site soiI sampling study are presented
in this paper.
The Atomic
Energy Commission’s
Nevada Test Site
(NTS)
lies approximately
sixty-five
miles
northwest
of
Las Vegas, Nevada
in the Great
Basin area. The soil on
and around
the Nevada Test Site is primarily
of volcanic
origin. The valleys are composed
of gently to moderately
sloping
alluvial
fans and
terraces.
The
soil is of coarse
textu{e
with
low organic
content
and low water-holding

89

characteristics.
The mountains
are steep to very steep and
composed
of
sedimentary,
metamorphic,
and
igneous
stone.z
This
soil
survey
was begun
at twenty
populated
locations
around
the Nevada Test Site and two unpopu-
lated
locations
(see
Fig.
1). These
locations
are both
inside
and
outside
the
fission
product
fallout
patterns
defined
for
the
test series above.
Baker, California
and
Kingman,
Arizona
were selected
as background
stations.
Initial
soil samples were collected
from profiles to deter-
mine
vertical
disposition
of
plutonium.
Two
proffle
samples
were collected
in the vicinity
of each location,
usually
three
to five miles apart. Profdes
of 23-cm deep
and 200-cm square were sampled with layers divided at 1,
3, 7,and
15 cm.
Since desert
soil is too dry and too
coarse
to use
cookie
cutter
or auger
sampling
methods,
the
samples
were collected
by a pit
technique.
A pit was dug as deep
as necessary
to
accommodate
the
maximum
sampling
depth
plus some working
room. One face of this hole was
left vertical.
From this face was trowelled
or scooped
the
desired
thickness
and
area
layers.
A
f~ed-size
scoop
works well. After the scoop is inserted,
its mouth
may be
covered
with
a broad
knife
to fix the sampled
area or
volume.
Also, it is convenient
to slide a flat plate under
the inserted
scoop to prevent
mixing any material
which
falls into
the sampled
area with
the subsequent
sample.

(’4

N
‘“[w

Fig. 1

After
the layer is removed,
surrounding
material
may be
cleared
away to prevent backfall
which may hinder sam-
pling the lower layers.
Aea
sampling
is done with a scoop technique.
Not
less than ten scoops totaUing more than
1 ft2 are used to
composite
one sample. As above, the scoop is designed for
a fixed
sample
depth
and
area. To date,
this has been
5 cm by 100 cmz. Based on the profile results, it appears
that 5-cm deep will be sufficient
for most cases.
All samples
are prepared
for analysis
in a similar
fashion.
The
sample
is first
screened
and
subdivided.
There
is general
agreement
that plutonium
will reside in
some fme fraction
of the soil. There is not agreement
of
what
fraction
to
eliminate.
Some
analysts
discard
the
material
more coarse than 200-mesh;
some discard mate-
rial more
coarse than
25-mesh.3
The SWRHL procedure
uses 10-mesh as a dividing point.
The more coarse mate-
rial is gently ground in a mortar
to break up the clods and
screened.
The fine fraction
is divided by a riffling appara-
tus to provide aliquots
for analysis. An aliquot sufficient-
ly small to be handled
in a 100 cc bottle (approx.
100 g)
is chosen
for plutonium
analysis
and another
aliquot
of
about
400 cc (about
500 g) is selected
for gamma spec-
troscopy.
The
small aliquot
is dried
at
110”C, ground
and
mixed. One-g aliquots
are ignited at 700”C and dissolved
in “a Teflon
beaker
by digestion
with
nitric
and hydro-
fluoric
acids. Nitrate,
fluoride,
and silica are removed
by
evaporation
to dryness followed by repeated
evaporations
in the
presence
of hydrochloric
acid.
Plutonium
is ab-
sorbed
from a 9~ hydrochloric
acid solution
of the resi-
due on a column
of AG 1 x 2 anionic
exchange
resin.
Co-adsorbed
iron
is removed
from
the resin with 7.2~
nitric acid after which the plutonium
is reductively
eluted
from
the
resin with
1.2~
hydrochloric
acid containing
0.6% hydrogen
peroxide.
The separated
plutonium
is elec-
trodeposited
from
l&l ammonium
sulfate
media
onto
stainless steel planchets.
The activity
of the
Iutonium
is
determined
by alpha spectroscopy
using 4
u as an in-
ternal reference standard.4’s
The 400 cc aliquot is counted
40 min on a 4 x 4 in.
NaI(Tl)
crystal
coupled
to a 400
channel
pulse-height
analyzer.
The
taped
spectrum
is analyzed
b
solution
for
‘“(k
‘loK JM:
~:trix
lEIW, ZMRa, 2’2~,
,
,
,
RU

and 9sZr.
Gamma-scan
results show nothing extraordinary
for
the locations
sampled for this survey.
Typical
results
which
have been found
for pluto-
nium are shown
in Table
I. The values shown
are com-
puted
from
the concentration
in pCi/g at the two-sigma
confidence
level. These results are preliminary
and subject
to minor
modifications
as procedures
are refined
or re-
peated analyses are performed.
Plutonium
was detected
in only
the top
3 cm in
most cases and the profiie
pairs agreed to within a factor
of three in most cases. Three area samples were collected
at Lathrop
Wells to evaluate
the variance within
a group
of cores and between
locations,
but
unfortunately
they
were
collected
at one of the disagreeing
cases. Another

.

90

TABLE I

MEASUREMENT
OF 239Pu IN SOILS
AROUND
THE NEVADA
TEST
SITE

Location
(see Figure 1)
(mC?@’
)

Penoyer
VAlley
1 mi E of Co Line
6mi
Eof
Co Line
130 * 6.0
6.7 I
1.5

Queen
City
Summit
profile
at summit
(1000
cm2
surface
scraping)
19 *
l.O
22 *
1.9

Highway
25/Reveille
Turnoff

Lathrop
Wells

5.7 f
(3.71

2mi
E
2rniw
2 mi E surface
2 mi E surface
2 mi W surface

6.1
* 0.94
().3 ~ 0.2
17 & 3.2
2.6
*
1.3
0

Alamo

Beatty

Tonopah

Warm Springs

Moapa

Diablo

Goldfield

Butler
Ranch

Caliente

IndianSprings

Furnace
Creek,
California

Scotty’s
Jet.

4.2
mi N
3.4
rni s

2.4 mi N
6.6
rni s
3.8
~ 0.88
3.() f
0084

1.5 mi S
3.9
mi NW
0.5
k 0.2
1.1 I
0.52

2.6” mi E
4.5
mi w

4mi
NW
7mi
Nw
1.9 * 0.92
0.6 * 0.4

2.4
mi N

2mis

7.8
+ 1.8
8.2 ?
1.3

3.4
mi s
3.8 mi N
4.3
f
0.79
2.5
? 0.58

1.9 mi S
2.3
mi N
1.4 k 0.89
22
? 2.4

4mi
N
4.2
mi S
1.1 * 0.64
0.8
* 0.3

1.5 mi E
3.7
mi w

0.9
*
0.5

1.5 * 0.39

1.3 mi S of Inn
0.6
mi N of Ranch

0.4
*
0.1

0.6
+ 0.3

2.3 mi S
2.1 mi N

4.0
*
1.3
1.2 f
0.39

Location (see Figure 1)

Clark
Station
lmiw
2mi
E

Hiko
3.6
mi N
1.5 mi s

Kingman,
Arizona
.
0.6
rri
E
1.6 mi W

Baker,
California
1 mi N of Airport
6 mi N of Airport

Death
Valley
Junction,
1.4 mi S
California
2.1
mi N

Las Vegas
3miw
5misw
.--.
-—. ——-—..
——————.——
———
———.
—

*This result is under question. Another sample wiU be analyzed,

sample
was coUected
from
a cultivated
field
in which
plutonium
was found,
however
no plutonium
was found
in the barley growing there. No data correlations
have yet
been made between
these data and data generated
during
the test periods when plutonium
was known to have been
released.
It is noteworthy
that
Lrrthrop
Wells was in or
near the fallout pattern
of many of the Hardtack,
Phase II
experiments
and Butler
Ranch
lay in the fallout
pattern
of the Smoky
Event of Operation
Plumbbob.
No analyses
capable of defining specific origins of the plutonium
have
yet been attempted.
The locations
sampled
in this survey
which showed plutonium
do coincide with fission product
fallout
patterns
defined
for
the
above
mentioned
test
series.
This preliminary
information
shows there is detect-
able ‘%
in the areas around
the Nevada Test Site and
point out four general areas for further
study. These areas
are
Lathrop
Wells,
Goldfield
to
Scotty’s
Junction,
Penoyer
Valley
to
Reveille Turnoff,
and Butler
Ranch.
The highest deposition
of ‘~u
is northeast
of the Nevada
Test Site with the second highest deposition
being south-
west
as
defined
by
this
survey.
Values
range
from
130 mCi/km2
to background.
Sampling
will now be ex-
panded
in these four areas to define distribution
patterns
as they now exist.

z%%
(mCi/km2 )

1.9 + 0.61
14 t
2.6*

1.6 f
().54
0.8 * 0.4

0.7
*
0.4
1.0 *
0.5

0.2:
0.2

4.() ~ 0.63
0.5
* 0.2

1.8 * 0.70
0.5
*
0.2

References

1. Samuel Glasstone cd., 77re Effects of Nuclear Weapons, U.S.

Atomic Energy Commission, (1962, rev. 1964).

2. Verr
Leavitt,
Soil Scientist,
Radiological
Research Program,
Southwestern
Radiological
Health
Laboratory,
Las Vegas,
Nevada.

3. Bortoli,
Gaglione, Natural and Fallout Radioactivity
in the
Soil, Health Phys., Vol. 17, pp. 701-710,
Pergammon Press,
New York, (1969).

4. N. A. Talvitie, Detertnination
of Plutonium in Environtnentai
and Bioiom”calSattwies by Ion Exchan.ce (as revised), South-

5.

western R;diologica-l Heal~h LaboratoryILas
Vegas, Ne&da.

N. A. Talvitie, “ElectrodePosition
of Plutonium and Thorium
from
Ammonium
Sulfate
Medium,”
ITR-19,
Southwestern
Radiological Health Laboratory, Las Vegas, Nevada (1969).

.

92

CONCENTRATIONS
OF PLUTONIUM,
COBALT,
AND
S1LVER
RADIONUCLIDES

IN SELECTED
PACIFIC
SEAWEEDS

by

K. M. Wong, V. F. Hedge, and T. R. Folsom

Scripps Institution of Oceanography

University of California, San Diego

La Jolla, California

ABSTRACT

Recent studies of marine organisms from the North Atlantic Ocean show
239pu have
been
found
in ~aweeds-
that exceptionally
high concentrations of

The
enrichment
of
other
radionuclides
also have been observed
in certain

speciesof seaplants in the Hudson River and in the Pacific. The high uptake of
radionuclides and the relative ease of sampling suggest that seaweeds may be

ideal for monitoring certain radio-activities in the marine environment.

For
this reason, we have initiated
a survey of
the concentrations
of

plutonium,
radioeobalt and radiosilver in several species of seaweedscollected
along the coastal water of Southern California. Preliminary findings concerning

the distribution of ‘9Pu
and some other radionuclides are reported.

Introduction

Recent
measurements
of
marine
samples
have
demonstrated
that
exceptionally
high concentrations
of
plutonium
are to be found
in seaweeds.1-4 lt is already
evident
that
the high
concentrations
in seaweeds
make
them
sensitive
indicators
of changes in plutonium
in the
environment,
and
that
relatively
small
samples
of sea-
weeds
that
are,
in many
cases, easy to collect
and can
easily be analyzed
with precision. Nevertheless,
the pluto-
nium
concentrations
in only a few of the various known
species of algae and marine
grasses have yet been meas-
ured and compared
with concentrations
in their environ-
mental
sea water.
For instance,
relatively
few measure-
ments
have been made
concerning
plutonium
in the red
algae and in the marine grasses living in relatively
uncon-
taminated
oceanic environment.
It is the purpose of this paper to report findings of a
preliminary
survey of plutonium
concentrations
in a few
selected
organisms
collected
recently
along the coast of
Southern
California.
The samples
include
several species
for which no previous studies have been made. Also, some
identical
species were collected
in several different
marine
environments
for
comparison
of
their
plutonium
con-
tents.
Wherever
possible,
correlations
have been
made

between
plutonium
concentrations
found
in the species
and the concentrations
of certain other nuclides that have
been found
useful in the past for monitoring
the progress
of
radioactive
contaminations
from
fallout
and
from
coastal
(and
shipboard)
reactors.
These
latter
include
‘Vo,
?0,
limAg,
and
a few other
gamma-emitting
nuclides.
It is apparent
that
many
of the seaweeds
may be
useful as monitors.
They are abundant;
several species are
widely
distributed;
they
usually
may
be collected
near
sources
of pollution,
reactor
discharge
pipes, and sewage
out-falls.
They
integrate
effects
of environmental
con-
taminations,
depending
upon
their life spans, for periods
of less than one year to more than 24 years.s
These
preliminary
results
tend
to emphasize
that
seaweeds
might
be still more useful if more were known
as to the rates by which trace elements
were accumulated
by the separate
genera and species, and also if more were
known about their responses in different
environments.

Methods

Twelve species of seaweed were collected
from five
stations
along the coast of Sout hem California
as shown

93

in Fig. 1. Certain
samples
of the same species were also
collected
from
different
stations
at different
times
to
check for variation
of plutonium
concentration
as a func-
tion of geographic
location
and collection
time. All these
samples
were
collected
between
December,
1970
and
July, 1971.
The
detailed
analytical
procedure
has been
fully
described
elsewhere.6
The
collected
samples
were sepa-
rated
and identified
by genera, then washed
in sea water
to
remove
sand
and
loose
foreign
materials.
The
wet
samples
were
weighed,
dried
at
10O”C, and
ashed
to
constant
weight at 450 to 475°C in a muffle furnace.
The
ashed
samples
were
dissolved
in HN03 -HC1 and
236Pu
tracer was added
to serve as radiochemical
yield monitor.
The plutonium
was separated
and purii7ed
by anion ex-
change
column,
electroplated
onto
a stainless steel disk,
and determined
by alpha spectrometry.

Results and Discussion

Table
1 summarizes
our data
on samples
collected
from
the
coastal
water
of
Southern
California.
The

23%
and
four
gamma
t?rnitk?rs
are
concentration
of
shown.
The samples
are grouped
separately
into red algae,
brown
algae, green algae, and two kinds of marine grasses
so that their behavior
can be discussed separately.
It appears,
from Table 1, that
there is a wide varia-
tion in 239pu concentration
among the different
species of

seaweeds.
Also,
variations
by factors
from
3 to 5 have
been observed
even when the same species were collected
at different
times or locations.
It may be noticed
first that the highest
concentra-
tion of 9sZr/9sNb are associated
with
all of the samples
that were collected
during
the period
of June 21 to July
4, 1971.
This suggests
that
a new source
of fallout
has
been encountered
this year.
The concentration
of ‘To,
“%20 and
llOmAg in the samples,
however,
do not corre-
late with
the same increase
in 239pu or gSZr/9sNb activ-

ities. Since
‘To,
~o
and
llOmAg are believed
to have
been
released
from the San Onofre
Nuclear
Power Reac-
tor (collection
site B in Fig. I), this negative
correlation
of ‘To
in these samples suggests that the new activity did
not come from the San Onofre effluents.
Further
examination
of the samples containing
the

1w
1$0.
., !!,.!
I;, rr
115.
I In’

I
I
i
I
f
1
I
1
1
I
r
I
\’
r
r
I
f
,
8
1
I
!
1
v
1
I

k

SIiv
—-.
rWWclsco
. .

LOIW.I

oE

--- -----

MEXICO

L

30” -

‘dS—

d
1
I
I
1
I
1
I
I
I
1
!
[
1
t
,
I
9
1
t
9
I
I
1
0
I
I

139”
130.
,z~.
1/u.
115.
Iw

Is.

\.
>

\#f
~;
~jo

COfOn0d08
‘:.
v

,

.

.

Y

1“

Rg.
1

94

TABLE
I

RADIONUCLIDE
CONCENTRATIONS

Collection
Sample

Red Algae
Gelidium
sp.
Geiidium
sp.
Gelidium
sp.
Amphiroa
sp.
Corallina
sp.

Brown
Algae
Macrocystis
sp.
Macrocystis
sp.
Macrocystis
sp.
Eisenia
sp.
Eisenia
sp.
Egregia
sp.
Egregia sp.
Zonaira
sp.
Zonaria
sp.
Sargassum
sp.
Sargassum
sp.
Dictyopteris
sp.

Green
Algae
Ulva sp.

Surf
Grass
Phyllospadix
Phyllospadix
Zostera sp.

Sp.
Sp.

Date

12-12-70
12-12-70
7- 1-71
6-30-71
3-30-71

Apr. 64
6-30-71
7-4-71
Apr. 64
7- 1-71
3-10-71
7- 1-71
3-1o-71
7- 1-71
6-30-71
7-4-71
6-30-71

6-30-71

3-19-71
6-21-71
6-30-71

Site

B
c
El
D
B

c
D
A
c
E
B
E
B
E
D
A
D

D

B
c
D

IN PACIFIC

239pu

0.58
* 0.07
0.42
* 0.04
2.20
* 0.15
2.10
* 0.20
1.48 * 0.15

0.71
* 0.06’
0.71
* 0.05
0.67
t
0.10
1.00
* 0.05’
2.8S
k 0.25
0.44
* 0.03
1.55 * 0.09
1.65 * 0.19
5.50
t
0.30
0.52
* 0.05
0.72
* 0.25
3.70
* 0.20

1.20
f
0.40

0.61
* 0.03
0.90
* 0.12
0.68
* 0.08

SEAWEEDS

dpm/kg wet samplea

‘co

98
<2
<2
..
950

..
<2
..
-.
..
49
<2
..
<2
..
..
<2

<2

960
<2

..

10

3

9
..

48

.-

<2
..

..

...

<2

<2

46

4
..

<2

6

61

18
..

tlomAg

12
<2
<2

41

..
<1
.-
.-

<1
<1
..
<1
..
..
<2

6

93
<2

aThe reported error for
Pu is one standard deviation of the counting data. The counting error for other radionuclides
is equal to or less than
10%. Activity
below detection
limit is indicated
by less than value.

bSee Figure
1 for sampling locations.

cData from Pillai et al., 19641

highest 23%
concentration
clearly shows that the greatest
increase in 9sZr activity
was related
to the sampling loca-
tion near Coronado
Island as shown
in Table H. This is
consistent
with
geographic
variations
found
in earlier
studies
of fallout
carried out in surface seawater
west of
California
in 1964- 1965.7
For example,
Table 111shows
that
the ‘~u
concentration
in seawater
increased
by a
factor
of 3 between
samples
collected
from
the Scripps
Pier and those
collected
10 miles from
the coast.
Since
Coronado
Island is about
8 miles from the coast, the 3 to
5 fold
increase
of m~u
concen~ation
in the seaweeds

‘Zr
%Nb

<4
<4
930
780
<4

.-
540
..
..
..
<4
290
.-
92
308
..
910

129

<4
411
200

collected
there
corresponds
with
the
expected
higher
plutonium
concentration
in the seawater
at this distance.
It appears
that for short
periods
after new global fallout
occurs,
there
is an upward
gradient
of fallout
concentra-
tions in the surface seawater
as one goes westward
from
the coast.
one
239pu measurement
of
sea water
from
‘ie

Scripps
Pier was made in June
1971. A concentration
of
0.16 * 0.04 dpm/ 100 liters
was found.
This is nearly
a
factor
of 2 higher
than the value found
in 1964. This is
also
in agreement
with
the
higher
23~u
concentration

95

.
I

++

v-l
0
0
u

m0
0

+1
m
m.
0

23
0

+1

~
0

+1
0
0)
0

ii
-H

0
0
++

.

.

“~“ijl
L9L9

96

TABLE
III

G

VARIATION
OF ‘%,
‘Sr
and *37CaIN SURFACE
SEA WATER
WEST OF CALIFORNIA
1964
(Data
from
Folsom
et al.,
1966)7

Station

Scripps
Pier
32”40’N
116”30’W
35° 12’N
120”57’W
34° 16’N 120”03’W
33”49’N
121”50’W
33”50’N
126”35’W
33”00’N
132”30’W
32”30’N
133°00’W
30”N
140”W

Miles
From
Coast

(0.2)
10
30
100
300
700
720
1,100

observed
in the seaweed from Coronado
Island, if, as we
believe, the plutonium
concentration
in the sea water still
increases seawardly
as it did in 1964. Using this new value
of’%
concentration
for the coastal water,
the concen-
tration
factor for the Southern
California
seaweeds range
from
260
to
3500.
These
values
fall within
the
data
obtained
by Pillail and Noshkirr.3
It is interesting
to note that the highest concentra-
tion
of
Z3~u
ever
found
was. in
the
North
Atlantic

Sargassum as shown
in Table IV. On the other
hand,
a
Pacific species of Sargassum was found
to be one of the
lowest concentrators
observed in the present study. It will

pCi/100 liters

23%
%r
137~

0.04
0.11
0.11
0.14
0.15
0.30
0.26
0.30

..
12-46
9.7
16
5.0
9
19
27
37
59
57
66
..
..
37
48

be noted
in Table
I that
another
brown
algae, Zonaria,
was the highest concentrator
found in this study but high
concentrations
were found in red and green algae and also
in the two marine grasses. This example
further
illustrates
how hard it is to generalize
about
the behavior
of pluto-
nium in the marine environment.

Conclusion and Future Work

The
results
obtained
so far from
this study
pose
more
questions
than
answers
concerning
the interaction

TABLE IV

PLUTONIUM CONCENTRATION
IN ATLANTIC
MARINE ORGANISMS
(Condensed
from data of Noshkin
et al., 1971)3

Sample

Blue Mussel
(Mytilus
edulis)
Blue Mussel (Mytilus
edulis)
Oyster
(Ostrea
virginica)
Scallop
(Pectem
irradians)
Scallop
(Pectem
irradians)
Scallop
(Pectem
irradians)
Starfish
(Asterias
forbesi)
Kelp
Staghorn
(Cadium
fragile)
Chondrus
crispus
Fucus
vessaculosis
Ascophyllum
nodisum
Sargasso
Weed (Sargassum
sp.)

Oman

body
shell
body
Adductor
muscle
body
sheU
body

‘%
range,
dpm/100
kg wet

36-97
89-98
19-31
2-7
78-131
115
167-220
20
39
76
139
126-301
124-18,
500

97

of the environmental
plutonium
and seaweeds.
As indi-
cated earlier, this is only a preiiminruy
study by which we
hoped to raise useful questions.
We may conclude
then, that all species of seaweeds
concentrate
plutonium
and that seaweeds may be a sensi-
tive indicator
for the detection
of variations
of plutonium
concentration
in the
marine
environment;
also, further
work
should
be done
to correlate
plutonium
concentra-
tion between
sea water and algae, and that a more com-
prehensive
survey
of the marine
environment
is needed.
By comparing
samples
collected
near the nuclear
plant
with samples of the same species on other
coastal collec-
tion sites no evidence of anomalous’%
was found near
the
plant,
and
definite
evidence
was found
that
“%0,
%o
and
‘lOmAg had been
coming from
the plant.
By
comparing
different
species
collected
near
the
nuclear
plant,
the red algae, Gelidium
and Cbrallina, and a surf
grass, Phyllospadix,
accumulate
higher concentrations
of
cobalt
and silver radionuclides
than did the brown
algae.
(It
is interesting
to note
that
one
species
of sea hare,
Aplysia californica,
that is believed to prefer red algae as
food also shows higher concentrations
of ‘~o,
60Co and
1lom Ag.
Typical
concentrations
were:
2200,
180,
260 dpm/kg wet sample respectively.)
Besides the accumulation
of more data from analy-
sis, we believe certain
controlled
experiments
should be
set up to study
the rate and mechanisms
of plutonium
uptake
by sea plants.
It appears
that
more experiments
similar to those
done
by Ward (1966)8
are necessary
to
establish
quantitative
relationships.
We hope
to
make
future contributions
in this area.

Acknowledgment

We are most grateful
for the valuable
help given by
James R. Stewart
of Scripps for the collection
and identi-
fication of many of the samples.
This
work
was
supported
by
the
U.S.
Atomic
Energy
Commission,
contract
No.
AT(04-3)-34,
P. A.
71-15,
and the U.S. Office
of Naval Research,
cent ract
No. USN NOO014-69-A-0200-601
1.

References

1.

2.

3.

4.

K. C. Pillai, R. C. Smith, and T. R. Folsom, “Plutonium
in the
Marine Environment.”
Nature 203:568-571,
1964.

K. M. Wong, J. C. Burke,
and V. T. Bowen, “Plutonium
Concentration
in Organisms
of the
Atlantic
Ocean:
Fifth
Annual Health Physics Society Mid-YearTopical Symposium,
Idaho FaUs, Idaho, November 1970.

V. E. Noshkin, V. T. Bowen, K. M. Wong, and J. C. Burke,
“Plutonium
in North
Atlantic
Ocean Organisms; Ecological
Relationships:’
Third Natioml
Symposium on Radioecology,
Oak Ridge, Term., May 1971.

A. Aarkrog, “Radioecological
Investigation of Plutonium 1ssan
Arctic Marine Environment:’
Health Phys., in press.

5. ‘Y. E. Dawson, Murine
Botany,
HoIt, Rinehart
and Winston,
Inc., N.Y., p. 73, 1966.

6. K. M. Wong, V. E. Noshkin, L. Surprenant,
and V. T. Bowen,
“Plutonium
in Some Organisms and Sediments;’
U.S. Atomic
Energy Comm. Rep. HASL-227: I-25-I-33, 1970.

7. T. R. Folsom,
K. C. PiUai, and T. M. Beasley, “Studies of
Background Radioactivity
Levels in the Marine Environment
near Southern California;’ S10 Ref. No. 67-7-A, B,C, 1966.

8. E. E. Ward, “Uptake
of Plutonium
by the Lobster, Homarus
vu~aris,”Nature 209:625-626,
1966.

.

.

98

RESUSPENSION
OF PLUTONIUM-239
IN THE VICINITY

OF ROCKY
FLATS

by

H. L. Volchok

Health and Safety Laboratory

U.S. Atomic Energy Commission

New York,
N.Y.

ABSTRACT

Continuous,
high-volume
airborne
particulate sampling has been main-

tained for
over a year, close to, and downwind
from,
the Rocky
Flats plant.

The sampler is in the vicinity of the highest ground concentrations of 23%
as

determined
in
a 1970
inventory.
The
concentrations
have averaged about

2 fCi/m3
of air sampled, 10 to 100 times higher than the expected levelsfrom
fallout.
In
addition
a qualitative correlation
is demonstrated
between wind

velocity and 239pu concen~ation
in the air.
The results to date su99est ‘resu-

spension factors of
between
10-7 and
10-9 depending
upon
the assumption

taken, for the depth of soil re-entrainment.

.. ... ........ ..._....- ....._- ___—.._.-- .._......

Introduction

FoUowing the May 11, 1969 fire at the Rocky Flats
plutonium
processing
plant,
and the publicity
generated
by Dr. Marten’s
demonstration
of plutonium
in the soil
off
the
plant
site,l
the
Health
and
Safety
Laboratory
(HASL)
undertook
a program
to study
the distribution
and inventoW
of ‘%
in the area. This study was com-
pleted
and
published
in August
1970.2
In summary,
it
was found
that
the
most
likely
source
of the
offsite
plutonium
in the
environment
was the
barrels
of con-
taminated
oil which
had
been
stored
on the southeast
corner
of the plant
property,
and which were known
to
have leaked. The pattern
of contamination
on the ground
was generally
compatible
with the average wind vectors in
that
region.
The upper
limit of the inventory
of offsite
‘%
attributable
to contamination
from
the plant
was
found
to be 5.8 curies. Figure 1, from the report by Krey
and
I-Iardy2
is a contour
representation
of
the
‘%
distribution
in the
Rocky
Flats area. The contours
are
lines of equal 23’% deposit in units of mCi/km2.
It seems
clear
from
Fig. 1, that
the highest
levels off the actual
plant
property
are predominantly
to the east and south-
east, with
the hot spot
as defined
by the contours,
just
adjacent
to the area where
the leaking
drums
had been
stored.

Since the available evidence suggested
that this off-
site plutonium
contamination
was not a result of the fire,
and could
be generally
correlated
with the average wind
patterns,
it seems reasonable
to assume that resuspension
and transportation
by the wind was responsible
for this
ground
deposit.
Hence,
in mid-1970
we set out to obtain
data
on resuspension
of plutonium
in the Rocky
Flats
area.

Sampling and Analysis

We started
with a single sampler
placed as close as
possible
to the area which
we believe to be the source.
l%is is the
so called
Pad,
where
the
barrels
had
been
stored.
The sampler
is a standard
HASL surface-air
pro-
gram, Roots
Connersville
blower.
Using 8-in .diam
Micro-
sorban
falter paper, we routinely
sample continuously
for
a week at an average flow rate of about
1 m3 /rein. Figure
2 shows the sampler
in a typical
louvered
housing on the
HASL
roof.
At
first
the
filters
were
composite
into
monthly
groups
for analysis,
then,
starting
in the
late
summer
of 1970,
weekly
samples
were analyzed.
All of
the
samples
have been
analyzed
for both
‘9/2’%% and
23%
under
contract
with
Trapelo
Division
West,
of
Ricl&ond
California.

99

‘-4,
,
.-L. - % .A--
-
,.
f
~
. .--_-y%-
.’

Fig. 1

Results and Discussion

Figure
3 illustrates
all of the weekly
data on ‘~u
concentrations
in the air near Rocky
Flats,
as a function
of time. The concentrations
vary over more than a factor
often,
in the weekly samples, with a low of about 0.3 and
a high of over 6 fCi/m3.
A smoothed
version of the data is
obtained
by averaging over each month,
as shown
by the
dashed
curve
in
Fig.
3.
Here,
there
appears
to
be a
downward
trend
through
the summer,
increasing
as the
samples get into fall and winter.
This is of course qualita-
tively correlatable
with
wind
intensities
at Rocky
Flats,
and more will be said about this in a later section.
In Fig. 4, the monthly
Rocky
Flats air concentra-
tions of %%
are compared
with similar data from other

sites in mid-latitudes
of the Northern
Hemisphere.
Clearly
the available results
from Denver, 3 New York City,4 and
Ispra,
Italys
are similar.
All three
exhibit
the expected
seasonal
variation
of
worldwide
fallout,
coming
down
from the spring-summer
peak, to a winter
low. The high-
est value at any of these
three
sites, in this period,
was
about
.13 fCi/m3.
The
Rocky
Flats
results
are
quite
obviously
greater
by more
than
a factor
of ten,
and as
mentioned,
indicate
an almost
opposite
seasonal
trend.
The rather
obvious
conclusion
from these graphs,
is that
the air near the Rocky
Flats plant is definitely
contamin-
ated, and that the concentrations
of plutonium
at this site
are a factor
of ten or more higher than one would expect
from worldwide
fallout.

.

.

100

Fig. 2

Note that on the scale of fCi/m3 of air, as shown on
the graph of Fig. 4, the maximum
permissible
concentra-
tion (mpc)
would
be 60 units.
This is the recommended
level soluble
plutonium,
with
bone
as the critical
organ,
for nonoccupational
exposure.
So, on the average, this air
at the edge of and downwind
of, the contaminated
area, is
running between
about
1 and 10% of the mpc.
Another
method
of showing
the probable
source of

~

?lutOnlm-a33
concmtrat
ion
in
Gir-
mar
Rocky
rhea

10 IY

9

ccljJ

1

0.5

0.1
i
1

‘ July
,
Auq.
t Sapt.‘ Oct., Mov
.
1
I
Me.
I

1970
am.
1971

the plutonium
in the air near Rocky
Flats is by use of the
ratio Zapu/Z~u.
~
Dr. Harley pointed
out in his Presen-

tation,
there
was a characteristic
238/239
ratio
in the
atmosphere
from
weapons
tests
prior
to
1965;
this was
about
0.03,
which
is
also
the
approximate
ratio
of
weapons-grade
plutonium.
However,
when the SNAP-9A
power
source
burned
up in the
atmosphere,
in
1964,
23~u was added
to materially
increase
enough
additional
the ratio
in surface
air from
1966
on. The ratios in the
Northern
Hemisphere
were
summarized
in Dr. Harley’s
Table
W showing
a peak
of about
0.5 in 1967, Table I
lists the most
recent
data
available
from
our surface
air
sites, Denver3
and Ispra,s
for comparison
with the results
on Rocky
Flats
samples.
It seems very clear, from these
values,
that
most
of the plutonium
in the air at Rocky
Flats
has about
one third
of the ratio of weapons-grade
plutonium
found
in
worldwide
fallout.
So,
again
the
evidence
strongly
suggests
that
for the most
part, pluto-
nium in surface
air near the plant,
is contamination,
that
the
source
is the
pad
area, just
west
of our
sampler.
Additionally,
the
surface
air
in Denver
appears
to be
uncontaminated
by plutonium
from
Rocky Flats, at least
to the degree of the sensitivity
of this ratio.
Perhaps
the
best
evidence
that
resuspension
is
playing
a major
part
in the elevated
plutonium
levels in
the
surface
air near
Rocky
Flats,
is the relationship
of
these
data
to the winds.
We have tried
to correlate
the
concentration
results
with
the
available
wind
data
ob-
tained
at Rocky
Flats, in numerous
ways. The problem
in
this sort
of exercise,
is that
the shortest
period
of our

J

0- QuaZt*rlYcmnver
data I

#
#
\o
UW
York City

\
\
\
\
\ G,

XmpZ~,Italy

TxauRJl4

PlutOniIm-239
Concantratlona
in
Glx-at
various
Gltaa
.01
Jun9- July
Auq.
Sept.” Ott.
I
mm.
,me.
aJan.
1970
AS71

Fig. 3
Fig. 4

101

TABLE I

‘8 Pu/=@u
IN SURFACE
AIR

(Mean
values for the last half of
1970)

Moosonee,
Ontario
0.08
New York
City
0.11

Sterling,
Virginia
0.10
Miami,
Florida
0.08
San Juan,
Puerto
Rico
0.08
Denver,
Colorado
0.09
Ispra,
Italy
0.08

Rocky
Flats,
Colorado
0.03

sampling
is one
week,
hence
we have
to
do a lot
of
averaging of the wind data, and this may tend to oblite-
rate or mask any correlation.
We have tried correlating
the
weekly
plutonium
concentrations
with
such
things
as
mean wind speed, peak gusts, mean weekly gusts, number
of hours
in the sampling
period
that
the wind exceeded
various
speeds,
etc.
Qualitatively,
most
of these
wind
parameters
indicate
some correlation
with the plutonium
in the air. For example,
Fig. 5 is a plot of concentration
data
vs. mean
wind
speed,
for
the
one week sampling
periods.
In this plot we have differentiated
between
sam-
ples collected
in the summer
(open
circles) and autumn
(solid
circles).
Here,
as noted
on the Figure,
there
is a
rather
strong
difference
in the correlation
between
the
summer
and fall data. The linear correlation
coefficient,

(r) is only 0.18 for the summer months,
indicating
little if
any
correlation,
while
the
fall data (r= 0.8) are highly
correlated.
This is not
easy to explain,
with the type of

16

14

12

4

0

G

FIGURE
5

MEAN WEEKLY
WIND
VELOCITY

vs.
PLUTONIUM
CONCENTRATION

o

0
G
G

G“
‘0
<
,

G 00’8
0
0July
- Sept.
1970
r
=
.18

“o
o
&
G 0ct.
-
Dec.
1970
r
=
.83

a
.
.
.
L
i?
3
4
5
6
“1

fCi
per
cubic
meter

Fig. 5

.

102

sample and wind input
available to us. It suggests to me,
that
in summer,
when
the winds are lower and less vari-
able, the resuspension
is probably
more directly related to
short term meteorological
variables, not as yet obvious to
us. In the
fall, however,
the
good correlation
between
wind and plutonium
concentration,
even on this basis of
average weekly
samples,
seems to be attributable
to the
higher
average wind speeds. We can almost recognize
the
existence
of a threshhold
at approximately
8 mph; only
one of the summer
samples
averaged
above 8, while all
but three
of those taken in autumn
were above 8. Since
even
the
summer
samples
are
substantially
above
the
fallout
levels,
it
seems
as though
there
may
be
two
mechanisms
involved
in the resuspension,
one operating
below
about
8 mph,
and
not obviously
correlated,
and
another
which results in good linear correlation,
at mean
winds above about 8 mph.
For
completeness,
I feel
I must
mention
resus-
pension factors,
although
I really question
the usefulness
of this concept
in context
of the Rocky
Flats situation.
These factors are derived by dividing the ground deposit,
into
the
air concentration,
with
care in the
choice
of
units.
For
these
types
of
radioactivity
resuspension
studies, we can use mCi/m3 over mCi/m2.
But, implicit in
the
use of these
factors
is the assumption
that
the air
concentrations
observed are derived or related
to the soil
concentrations.
Since, in the area of our air sampling, the
gradients
in soil concentration
were very steep,
and we
did in fact find substantial
penetration
of plutonium
into
the soil, the simple use of the resuspension
factors
is of
doubtful
value. At any rate, at the site of our air sampler,
using the soil data reported
in HASL2352
a resuspension
factor
of about
10-9 m-l
was calculated.
In another
ex-
periment,
which will not be completely
discussed here, we
pressed sticky
paper to the soil surface, assuming that the
fme particles
which were retained
might approximate
the
readily
resuspendable
portion
of the soil and plutonium.
Using the results from this sample as the denominator,
the
resuspension
factor approached
10-6. Both of these values
are in the range of resuspension
factors
reported
earlier
from both experimental
and theroetical
considerations.

We have begun
studies
on the particle
size of the
airborne
particulate,
near the Rocky
Flats plant. The size
of the plutonium
particles
and whether
or not they are
attached
to
larger host
particles
are
critical
factors
in
finally
determining
whether
or not
this observed
resus-
pension
is a potential
hazard to man. The initial work has
been carried out under
contract
with Trapelo/West.
Very
preliminary
results suggest that the equivalent altimeter of
the PU02 particles
averaged less than 0.2 pm. We believe,
at this time,
that
the plutonium
is associated
with host
particles of median diameter
about
10 gm.
Our present
plans are to continue
the air sampling
at Rocky
Flats, expanding
to a few additional
pumps, to
define
the
downwind
gradient
of plutonium
in the air,
and to establish
some data in the northern
and southern
directions.
The studies of particle
size of the resuspended
plutonium
will be continued
and refined as some new and
better equipment
becomes available.

References

1.

2.

3.

4.

5.

“Report
on the Dow Rocky Flats Fire: Implications of Pluto-
nium Releases
to the Public Health and Safety,”
Colorado
Committee
for Environmental
Information,
Subcommittee
on
Rocky Flats, Boulder, Colorado, January 13, 1970.

P. W. Krey and E. P. Hardy, Jr., “Plutonium
in Soil Around the
Rocky
Flats
Plant:’
USAEC Report
HASL-235,
August
1,
1971.

“Plutonium
in
Airborne
Particulate,”
Radiological
Health
Data and Reports, Vol. 12, No. 6, June 1971, p. 335, EPA -
Office of Radiation Programs.

H. L. Volchok and M. T. Kleinman, “Radionuclides
and Lead
in Surface Air:
USAEC Report HASL-243 (Appendix), July 1,
1971, p.c-l.

EURATOM Joint Nuclear Research Centre, ISPRA Establish-
ment,
Quarterly
Report,
Reproduced
in
USAEC
Report
HASL-239, January 1, 1971, p. 111-40.

103

.

I

LOG-NORMAL
ANALYSIS
OF DATA
FOR
PLUTONIUM
IN THE
OUTDOORS

by

D. E. Michels

Dow Chemical Co.

Rocky Flats Division

Golden, Colorado

ABSTRACT

Detected amounts of
plutonium
are distributed
log-normally
for
most

groups of samples. When data are plotted on probability
paper, sharp distinc-

tions
may
sometimes
be
made
between
the
background
distribution
and
increments from a local source.

Because the detected amounts of plutonium
are not distributed normal-

ly,
arithmetical
averaging of detected amounts
is not valid. Similarly,
com-

posite
samples from
large areas yield
analyzed
values which
cannot
be
interpreted.
Additionally,
the proper standard deviation for background
sam-

ples refers to
a ratio
of
concentrations
rather
than
to
an
increment
as is
commonty reported.

Introduction

Since starting
to deal with data about plutonium
in
the outdoors
I have lamented
both
the variability
of the
data and the paucity of precise conclusions
that have .been
offered
concerning
the
distribution
of
plutonium.
Of
course, part of that variability
results from the nature
of
the dispersion.
Not only must we live with that but it is
the very thing
we must
describe.
One tool
that
so far
seems very powerful
in handling
plutonium
data is prob-
ability
paper.
Today
I wish to explain
the technique,
to
demonstrate
how it is applied
to real data, and most of
all, to
show
that
the
data
truly
can
support
concise
conclusions.

Discussion

First,
let’s look at some alternative
ways of plotting
the data while using a statistical
point of view. Any group
of data will have an average value and a degree of varia-
tion.
But we may have to search a little to find the best
way
of quantifying
both
the average and the variation.
This first slide shows four ways to describe the same data,
but
the
four
are not
equally
useful.
The
graph
in the
upper
left
represents
the
analytical
data
for plutonium

that we have to deal with. The data contain
an excess of
large values over what
a normal
distribution
would
con-
tain. Actually,
non-normal
distributions
for the analytical
values
should
be expected
for trace materials
anywhere
since zero concentration
is an impossible boundary.
Clear-
ly, when a one-sigma
or a two-sigma
distance
from
the
average
value
turns
out
to be a negative
concentration,
our point
of view should receive some serious adjustment.
If
the
data
are
truly
homogeneous,
then
some
mathematical
transformation
exists
for which
the trans-
formed
values
are
distributed
normally.
Finding
that
proper
transformation
is essential.
The graph in the lower
left (Fig. 1) corresponds
to the data after a proper trans-
formation
has been made. Thus transformed,
the data are
distributed
normally
and
then
(but
only
then)
do our
notions
about
averages and standard
deviations
become
appropriate.
Trying
to plot
a Gaussian bell-shaped
curve
from
empirical
data
is expensive
since several
tens
to
hundreds
of data are required
for any kind of precision
in
locating
the
actual
position
of the curve. However,
by
adjusting
the scales of our plots
we can get along with
fewer data. The graph in the lower right (Fig. 1) involves
cumulative
percent
and
a few tens
of data points
will
define
it nicely,
although
its curved
shape leaves much
room for gentlemanly
disagreements
about whether
devia-
tions from a true sigmoid shape may be meaningful.

40

--
IA

I.0I
PROBABILITY
/

30”

20-

lo-

ll K 0“7)
‘=FK7°

>
u
“’/-’--

10
305070
90 /
99

-3
-2
-i
o
+i
+2
+3

I
I

30

20

>
7
-0.5
0
0.5
1.0
1.5
-~.5.
020’
406080
100

Fig. I

The graph on the upper right (Fig. 1) is the kind I
wish
to focus on throughout
the rest of this talk.
It is
derived from the lower right graph by replacing the cumu-
lative percent
axis with a probability
scale. The probabil-
ity scale is one which is linear in units of standard
devia-
tion rather than in units of cumulative
percent.
There are four very considerable
advantages in using
this kind of plot.
First,
the plot will be linear when the
transformation
of the data does yield a standard
distribu-
tion.
Second,
the
question
of linearity
may
become
answerable
with as few as ten or twelve data points
(and
with twenty
data points one can acquire some real confi-
dence). Third,
the mean value for the data is given by the
zero-sigma
intercept,
which lies in the middle of the array
of plotted
points.
Fourth,
the slope of the array
is the
standard
deviation.
Primarily,
this plot
is a test
for as-
sumptions
we make about
the data. The linearity
checks
whether
we
and
for
the
directly
the
mean
value,

106

made
good
choices
for the transformation
distribution
type.
If linear,
the
plot
gives
two
most
important
statistical
parameters,
and
standard
deviation.
Some
convenient

graph
papers
are
available
commercially
which
have
a
normal
(Gaussian)
probability
scale along the horizontal
axis,
the
vertical
axis is variously
linear
or logarithmic.
Exotic (non-Gaussian)
distributions
as well as normal ones
are conceivable,
and
this
technique
applies
to them
ail
with equal validity.
Our job is to find the combinatio~
of
distribution
type
and data
transformation
which yield a
straight
line array.
Many sets of geochemical
data have
been found
to yield linear plots when a logarithmic
trans-
formation
is combined
with a Gaussian
probability
scale
and that
is the combination
I will discuss hereafter.
The
distribution
is commonly
called log-normal.
Before
we go further
into
log-normal
plotting
of
data, I want to introduce
a second concept
which also can
be answered
by graphical
techniques.
Multiple
sources of
plutonium
result in overlapping
distribution
patterns
and
part of our job is to find the limits of the overlap. Local
sources
like
Rocky
Flats
and
Los
Alamos
are
super-
imposed
on
the
world-wide
fallout
pattern,
but
the
world-wide
pattern
is itself
a composite.
In order
to
describe
accurately
the
geographic
limits
of
local

.

.

.

.

.

.

contamination,
as well as to take inventory
of the pluto-
nium
we need
methods
which
can
clearly
distinguish
superimposed
distributions.
Graphical
methods
are pre-
ferred
for
this
last
task
since
describing
the
edges of
anomalous
areas will involve subjective
decisions.
Again,
probability
plots are useful. Let’s look at an example.
The
data
shown
in Fig. 2, are from
Health
and
Safety
Laboratory
(HASL)
Report
235, and involve 33
soil samples taken in the Denver area. First,
the data are
arranged
in rank order
and a percentile
is computed
for
each datum.
When the plotting
is complete
we see two
distinct
legs and
conclude
either
that
the data are not
distributed
log-normally
or that
they
are not
a homo-
geneous coUection.
But we don’t really expect the data to
be homogeneous
anyway
since
the
reason
the
samples
were taken in the first place was to fmd out how much of
an effect
Rocky
Flats was having on the plutonium
in-
ventory neal Denver.
From
the
plot
we see that
the two legs intersect
near the value 3.0 mCi/krn2.
Using the 3.0 mCi/km2
value
as a criterion,
the
33 data
can be segregated
into
two

500
300

100

50
30

10

5
3

1

(unequal)
sub-groups,
each
of which
can be tested
for
homogeniety
by replotting
as two independent
distribu-
tions.
The
linear
plots
in Fig. 3 affirm
both
that
the
log-normal
plot is appropriate
and the the two groups are
homogeneous.
From
the
straight
lines we can estimate
mean
values and
standard
deviations.
Additionally,
the
successful
separation
of the
bulk
data tells us that
the
value 3.0 mCi/km2
is a good boundary
contour
for the
Rocky
Flats anomaly.
Only a small portion
of contamin-
ated samples would
show a value that low and the back-
ground values have a fair chance of being that high.
An important
element
of this kind of data analysis
is to treat
the data as groups not as individuals.
Indeed,
any single analytical
value should be considered
as with-
out meaning
when by itself. All meaning
comes from the
relationships
among
values.
Thus,
distributions
are the
primary objects to be described.
Making transformations
of analytical
data is an un-
common
pr?ctice
apparently,
but
it is both
useful
and
valid. For broaclly distributed
groups of data such as the
higher
content
sub-group
of the
HASL
data,
and
also

I
I
8
I
I
I
I
I
I
a
I
/1

SWW?IMPOSED
/

o
0
DISTRIBUTIONS

SAMPLE
33
}8
27
. .
--
7
.8
6

VALUE
PERCENTILE
1.8
2.0
2.0
.-

;jj

1950

S*O
6.1
9. I
--

97.;
97.0

/

1
I
I
I
i
i
1
1
1
1
I
1

5
10
20
30
50
70
80
90
95
CUMULATIVE % OF SAMPLES

Fig. 2

107

z

1000

500
300

100

50
30

10

5
3

I

I
I
I
s
1
m
I
I
I
I
I

A

A

TVJO
CONSTITUENT
DISTRIBUTIONS

A

A

A

A

I
1
I
I
t
I
1
t
I
I
I

5
10
20
30
50
?’0 80
90
95
CUMULATIVE % OF SAMPLES
Iyg.
3

others
that
I have studied.
it can be shown
that treatirw

the
analytical
data
as tzorrkzlly disfi”buted
is simply
no;
valid.
Therefore,
conclusions
based
on
averaging
the an-
alytical
values can be seriously
in error.
Which
transformation
of
the
data
is best
can
be
determined
only
by trial
and
error.
Logically,
we cannot
prove
that
any
transformation
is proper,
but we can show
when
a particular
transformation
is adequate.
Similarly,
we can show
that
making
no transformation
is sometimes
inadequate.
Figure
4 shows the results from testing eight
sets
of
plutonium
data
and
one
set
of %r
data
for
distribution
type, arithmetic
or logarithmic.
A W-test was
used
to estimate
the
probability
that
the assumed
dis-
tribution
is adequate.
In all cases the groups of data show
a high probability
of being log-normally
distributed.
In
half of the sets the arithmetic-normal
distribution
also has
a high probability
of being correct,
but in the remaining
four
cases a presumption
of arithmetic-normal
distribu-
tion
is not
warranted.
The
presumption
of log-normal
distribution
is never a bad presumption
and is never worse
than
the
presumption
of arithmetic-normal.
Often
it is

PROBABILITIES
OF DISTRIBUTION
TYPES

Data
P(normal)

Denver
Fallout
Italian
‘Sr
Denver
Background
(HASL)
Santa
Fe Background
Rocky
Flats
Anomaly
(HASL)
Italian
Pu
Rocky
Flats
Anomaly
(CCEI)
Denver
Air

0.099
0.74

0.78
0.17

0.059
0.032
0.000001
0.46

P(Log-normal)

0.91
0.90

0.78
0.75

0.63
0.54

0.45
0.39

.

.

.

108

much better,
so that in cases where we do not know what
distribution
type
actually
exists
presuming
a log-normal
distribution
is a good strategy.
When the data are quite
variable a (logarithmic)
transformation
is deftitely
neces-
sary.
When data
are transformed
we should
be careful
about
our interpretation
of the term average. The mean
values
indicated
by
the
50$Z0 intercepts
are
called
geometric-mean
values. They correlate
with the analytical
values in that
the geometric
mean is the antilog
of the
arithmetic
mean of the logarithms
of the analytical
values.
Thus, we should
take logarithms
before
taking averages.
For
the
case of log-normal
distributions,
we also
should carefully
examine
our interpretation
of the stand-
ard
deviation.
The
unit
of slope
in a log-normal
plot
involves
a
logarithmic
increment.
Thus,
the
standard
deviation
is a multiplier
of the geometric-mean
value. It is
not an increment
of the analytical values.
One
more
important
point
concerns
the
slopes

(SLOPE)2
= 02
= 02
+ UZ +
&

Noise
Signal

Fig. 5.
Components
of standard deviation.

associated
with log-normal
plots.
The slope is related
to
the components
of variance as shown in Fig. 5. Since the
variances are additive,
large variance
in either sampling or
analysis may mask the variance of the data.
An example
in which this problem
seems to exist is
from data reported
by Colorado
Committee
for Environ-
mental
Information
(CCEI)
(Fig.
6). When plotted
on
log-probability
paper the data yield a single linear m-ray of
high slope. The singularity
of slope suggests that the data
are
homogeneous.
But
geographically
the
data
involve
areas at Rocky Flats which lie both inside and outside the
contaminated
area. Hence,
the CCEI data should be ex-
pected to show two legs just as HASL data did.
Why does the CCEI data not
demonstrate
a local
increment
of plutonium?
Plotting
it together
(Fig. 7) with
the HASL data suggests that
the variances
due to analy-
tical and sampling
problems
may have masked the funda-
mental
variances.
The shallow
samples
of CCEI yielded
only
60% as much
plutonium
as did
the
HASL
back-
ground
samples.
The
CCEI
standard
deviation
is more
than
nine times as large as that
for HASL background,
and about
the same as the HASL standard
deviation
for
the Rocky Flats anomaly.
It would seem that the variance
associated
with
the CCEI data is too large to resolve the
underlying
variance of the background.
Consequently,
the
data
fail to
demonstrate
existence
of a local source
of
plutonium.

3.0 “
PLUTONIUM
IN SOIL
NEAR
1.0-
0
ROCKY FLATS
e

0.5 -
0.5 -

8
0.1 -

0.05 -

0.03 -

CCEI
1-13-70
0.01
,
1
,
t
,
,
,
*
,
1
5.
10
20
30
70
80
90
95
CUMULAT1VE5:
OF SAMPLES

Fig 6

109

&loot
I
I
I
I
k
s
I
1
I
I
I

t-W 50 -
PLUTONIUM
~
30 .
IN SOILS
NEAR
C
ROCKY FLATS

w
~
10 “
o

Q

z
5 -
~3
-
w

L
~
1.0-

w

o 5)
E“”
/“
o
0

S 0.11
I
1
I
s
1
I
t
1
I
1
I
5
10
20
30
50
70
80
90
95
CUMULATIVE % OF SAMPLES

Fig. 7

To summarize,
(Fig. 8) I have tried to show how a
graphical
technique
can be used to unscramble
and thus
help
interpret
data
about
plutonium
in the
outdoors.
Most
groups
of data
fit log-normal
distributions
better
than arithmetic
distributions.
Additionally,
when data are
plotted
on log-probability
paper one can decide whether
the data come from a single distribution
or from overlap-
ping distributions.
A graphical
method
is preferred
for
unscrambling
distributions
which overlap.
ampe
or me plots IS a icey pararuc(cl,
out me slope
can
become
so inflated
that
the
analytical
values
are

useless.
Sampling
and
sample-splitting,
particularly,
are
sources
of variance
more important
than
analytical
dif-
ficulties.
When a logarithmic
transformation
is appropriate,
the proper
mean value of the data is a geometric
mean
and
the corresponding
standard
deviation
has the prop-
erty of being a multiplier
rather than an increment.
Although
the
data
on plutonium
in the outdoors
tend to range greatly, the data often can support
interpre-
tations
that
are more precise than many reported
so far.

.

110

5C
3(

Ic

G%,

7L

I.(

0.:
0.:

0.

PLUTONIUM
NEAR
/
o
ROCKY
FLATS

/

o

i
I
a
1
s
*
I
1
#
1

5
10
20
30
50
70
80
90
95
CUMULATIVE % OF SAMPLES

GRAPHS
ARE
GOOD

LOG-NORMAL
FITS

WATCH
OUT
FOR VARIANCE

MORE
PRECISE
STATEMENTS
ARE
POSSIBLE:

fig. 8.

.

.

.

SOME
THOUGHTS
ON PLUTONIUM
IN SOI LS

by

J. W. Healy

Los Alamos Scientific Laboratory

University of California

Los Alamos, New Mexico

ABSTRACT

The resuspension of particles by wind or mechanical disturbance is one
of the major routes of potential
intake from plutonium
in soils, The actual air

concentrations
resulting from
resuspension depend
upon
many
variables in-
cluding the characteristics of the source, the degree of disturbance, the nature
of
the terrain,
and
the
meteorological
dispersion
and
deposition
processes

operating. Although
little data are available to characterize these variables and

to provide a general solution, some of the factors involved are discussed.

.-.. - ........—.. __...
—.__..
..--.
—... _._. -- .....

The
title
of this paper
took
very careful
negotia-
tions with the sponsors
of this symposium.
I lost on only
one
point -- I wanted
to
include
the
words
“Random
Thoughts”
since this would have given me complete
free-
dom to discuss almost any subject. However, upon further
consideration,
I fiid
the title
to be a little embarrassing
since it is very broad and, at the same time, it implies that
I might have some worthwhile
thoughts
to convey.
-
I would like to direct my remarks
toward
a few of
the factors which seem ‘to be of importance
in the resus-
pension
of materials
on the ground
in order
to permit
focusing
on the types of experimental
data and environ-
mental
measurements
which
are needed.
This is doubly
important
for plutonium
since the current
evidence
indi-
cates that the other major mechanisms
for intake,
such as
ingestion
or reconcentration
through
the food chain, do
not play as vital a role with plutonium
as with many of
the
other
isotopes.
Thus,
inhalation
has been,
and still
seems
to be,
the
mode
of intake
of importance
when
considering
plutonium
in the environs.
Before discussing
the normal
concept
of resuspen-
sion as a mechanism
to produce general air concentrations
in a region, we should consider
other
implications
of the
potential
for inhalation.
Thus,
entry
into an area having
plutonium
in the soil can result in a transfer
of some of
the material
to the body or clothing.
Later movements
or
removal
of the
clothing
with
subsequent
handling
can
result
in
some
of
this
material
becoming
airborne
to
produce
localized
air
concentrations.
Studies
with

contaminated
clothing
have
indicated
that
significant
transfer
can be accounted
for,l
although
the first step,
transfer from the surfaces to the clothing, has been poorly
invest igated.
Similar
mechanisms
can
occur
with
other
objects
such as tools,
or even the family pet, which are
taken
into
the
area.
While not
of primary
concern
in
dealing with
the safety
of people,
we must consider
the
possible intake by grazing or burrowing
animals since they
are more closely tied to the soil than man and could have
significant
intake
through
this close association.
Present
evidence
indicates
that
this is not a problem
in transmit-
ting the plutonium
to man because
the uptake
in organs
used for food is small and the uptake
from the GI tract of
man
is also small so that
these
two
factors
provide
a
strong discriminate ion against the plutonium
in soils. These
possible
intake
mechanisms
are subject
to many
of the
same variables as those to be discussed in the resuspension
mechanisms
and are mentioned
at this point to remind us
that
we must
consider
all possible
sources
of inhalation
and not concentrate
exclusively
on the single mechanism
as I shall do through
the remainder
of this paper.
Resuspension
and the general air concentration
re-
sulting therefrom
are very complicated
phenomena
which
will vary widely
depending
upon
the nature
of the con-
taminant
(such as particle
size), the characteristics
of the
surface
or the
soils involved,
the terrain
and vegetative
cover, and the particular
meteorological
conditions
at any
time.
Most studies of this process with radioactive
mate-
rials have
used
a simplifying
concept
of a resuspension

factor
in expressing
the results. This factor
is the ratio of
the air concentration
at a given location
to the quantity
of material
per unit area on the ground
at that location
and has been measured
under conditions
of normal
wind
actions
as well as with
added
mechanical
disturbance.
While this concept
can be useful in defined circumstances,
it gives
little
insight
into
the
nature
of the
processes
involved so that it is difficult
to apply this knowledge
to
other areas or forms of contaminant.
For example,
it does
not
account
for
the
size
of
the
area
or the
possible
existence
of
more
highly
contaminated
areas
upwind.
Estimates
of the dispersion
and deposition
characteristics
of material
from a uniform
source emitting
to the atmos-
phere indicate that significant
concentrations
of respirable
size particles can originate
from miles away. The resuspen-
sion factor
does not account
for dispersion
by the atmosp-
here
or for changes in the rate of resuspension
with, for
example,
wind speed or changes in atmospheric
stability.
A different,
and somewhat
more complex,
approach
is to consider
the mechanisms
of resuspension
separately
from
those
of deposition
and
dispersion
in the atmos-
phere.
In this way
each
point
of the area can be con-
sidered as a source of airborne
material and the concentra-
tion at any point
downwind
can be calculated
by use of
the correlations
derived from atmospheric
dispersion
and
deposition
studies
and
by integration
over
the
area of
deposition.
Similarly,
the magnitude
of the pickup
rate
(or fraction
resuspended
per unit time) can be studied by
measuring
the concentration
downwind
from a source on
the ground
under
various
conditions
of natural
or artifi-
cial disturbance.
This approach
is certainly
not as simple as that of
the resuspension
factor,
but by carrying out the measure-
ments
in such
a way
as to
gain
information
on
the
characteristics
of the
source,
the
meteorological
condi-
tions
and the airborne
concentrations,
one can account
for many
of the variables
and from
these
make an esti-
mate of the resuspension
concentrations
which will occur
for
different
areas
in which
the
size,
distribution
of
material
and
particle
size may
differ.
It must
also be
admitted
that, at the present time, there seems to be little
quantitative
data in the literature
which would permit the
making
of reliable
estimates
under
any condition.
Fur-
ther,
there
are processes
which
operate
over relatively
small areas, such as the small whirlwinds
frequently
en-
countered
in
desert
country,
which
could
provide
a
separate
source
of
resuspension
which
would
not
be
adequately
covered by a more general large area study.
The
work
of
the
soil
scientists,
particularly
Bagnold2
and
Chepil~-7
have given considerable
insight
into the mechanisms
of movement
by winds, particularly
under
conditions
of gross movement
such as occurs
at
high wind speeds over desert sands or plowed fields. Their
observation
of a threshold
velocity
of the wind speed for
this
type
of movement
is widely
recognized
as is their
demonstration
of the stability
of fine powders of uniform
particle
size even under relatively high wind speeds. How-
ever, it is not clear that these observations
are completely

applicable
to
the
problem
of concern
here, where
rela-
tively low concentrations
moving as suspended
materials
are of interest.
For example,
some observations
have been
made
of
air
concentrations
of
zinc
sulphide
particles
downwind
from
a single source
on the ground
at wind
speeds as low as 1.3 m/see. 8 At least we should design our
experiments
and measurements
to indicate the validity of
such concepts.
The
question
of the
behaviour
with
time
of the
deposited
material
has many
practical
aspects
but
few
answers.
For example,
aggregation
of the deposited
parti-
cles with soil particles
will result in differences
in behav-
iour
depending
upon
the soil particle
sizes, degree
of
natural
aggregation
and
the
stability
of
the aggregate
under
the
disturbances
expected.
We cannot
expect
a
permanent
f~ation
on
soil
particles
since
Chepil
has
noted
that
there
is a continuous
production
of small
particles,
at least in agricultural soils, under
the influence
of erosive forces,
but
the net effect of such aggregation
may well decrease
the overall susceptibility
of originally
fine
particles
to
movement
into
the
atmosphere.
The
gradual movement
of the deposited
material
into the soil
profde
by washing or alternate
freezing and thawing
will
decrease
the
surface
layers
which
are most
subject
to
disturbance.
Seasonal
variations
in vegetative cover, mois-
ture and even in meteorologfca.1 conditions
will affect the
possibility
of resuspension.
One can
visualize,
for
this
purpose,
two
limiting
conditions.
The first corresponds
to a fresh deposit where
the material
is exposed
on the surface of the ground and
other
surfaces
with
a particle
size distribution
character-
istic of the deposited
material and independent
of the size
distribution
of the soil particles.
Under these conditions,
the
deposited
material
is readily
available
and
can be
described
in terms
of the quantity
per unit area. In the
second
limiting
condition,
the
deposited
material
has
weathered
and
become
intimately
associated
with,
at
least, the top layers of the soil profile perhaps even to the
extent
of having
similar effective
particle
sizes through
the processes
of aggregation,
erosion,
etc. Here, only the
top layer is subject to resuspension,
with the definition
of
the
top
layer
dependent
on
the
degree
of mechanical
disturbance
or, perhaps,
the
wind
speed
under
natural
conditions.
In this case the amount
resuspended
is closely
related
to
the
natural
dust
from
the
surface
and
the
concentration
in the
soil
would
seem
to
be of most
interest.
Following
a deposition
we would expect a transi-
tion period from the fwst limiting condition
to the second
over a period
of time along with spreading
over a larger
area due primarily
to the processes
of surface creep and
sahation
in barren
areas but also to redeposition
of smal-
ler particles
in vegetated
regions. The
time required
for
this
transition
is indeterminate
and
probably
depends
upon
the
characteristics
of
the
individual
area.
Some
measurements
have been made in arid regions which indi-
cate
that
the
initial
air concentrations
decrease
with a
half-life
of about
one to two months,
however,
it is not
clear
that
these correlations
are not partially
associated

.

.

114

with
other
factors
such
as seasonal
variations
in wind
direction,
velocities
or
stabilities.
At
the
moment,
it
appears
likely
that
a decrease
in the
resuspension
will
occur as a deposit
ages but data do not seem adequate
to
characterize
the rate of decrease
or the time to attain
a
final,
reasonably
steady
state,
particularly
when differ-
ences in soils, terrain,
vegetation,
etc. from one region to
another are considered.
The above considerations
also bear on the question
of how we should measure and report
the data. There is
much historical
precedent
for the quantity
per unit area,
such as pCi/m2
and this seems appropriate
for the initial
period
following
deposition.
However,
when we sample
the soils for analysis, the result is measured
in concentra-
tion units
such as #Ci/g.
Eric Fowler
and his soils com-
mittee
at the Nevada Test Site have suggested a standardiz-
ation
of terminology
whereby
the results are reported
as
quantity
per unit area with a specification
of the depth of
the profile sampled
and, if possible, a specification
of the
soil density.
From these data it is possible to convert from
one to another.
For purposes
of considering
resuspension
we are primarily interested
in the top layer containing
the
contaminant
and
subject
to disturbance.
For
practical
reasons of sampling, it appears difficult to consider a layer
less than about
one centimeter.
If we could, again, stand-
ardize on some such thickness,
then the results would be
meaningful
in most
cases and the decrease
with time as
th~ material
penetrates
into the soil could be considered
in the studies of rate of resuspension.
While on the subject
of units,
I would
like to make a personal
plea for some
consistency
in methods
of reporting.
We see air concentra-
tions
reported
in pCi/cc,
fCi/m3,
aCi/m3,
etc.
While I
realize that
this is convenient
for the author
because
of
the lack of an exponent,
I have considerable
difficulty
in
making
the necessary
conversions
to compare
with other
papers
or with
the standards,
and I suspect
that
a few
errors
creep
into
the conclusions
of other
people from
such mental conversions.
It would, therefore,
seem worth-
while to consider reporting
our results in the same units as
the standards.
There
is one
other
consideration
in the measure-
ment of soils as connected
to resuspension
which I would
like to mention.
This is the fact that
the processes
in-
volved tend to average the pickup
from a relatively
wide
area. Thus, the real need in describing the ground deposi-
tion is not a point-to-point
sampling but, rather, averaging
over a significant
area.
This,
of course,
will affect
the
sampling strategy
although
I am certainly
not prepared
to
fully define optimum
area of sample size.

Finally,
we have considered
a few of the difficulties
of relating air concentrations
to soil concentrations.
There
are many more including
the problem
of soil drifting due
to eddies, redeposition
on a hard surface, and defining the
characteristics
of an actual
source.
While I believe
that
further
studies
of resuspension
mechanisms
are necessary
to further
define
and control
potential
problems,
I also
question
whether
the quantity
of material
deposited
is a
useful
parameter
for
control
purposes.
In view of the
many
variables
involved
in the
resuspension
process,
it
would
seem that
direct
measurements
of air concentra-
tions would
provide
more
direct
and useful information
than an equivalent
amount
of effort on soil measurement
followed
by
extrapolation
with
many
variables
to air
concentrations.

References

1.

2.

3.

4.

5.

6.

7.

8.

R. Butterworth
and J. K. Donoghue, “Contribution
of Activity
Released
from
Protective
Clothing
to
Air
Contamination
Measured
by
Personal
SampIersY
Health
Physics
18, 4,
319-323, (APril, 1970).

R.
A. Bagnold,
“The
Physics of Blown
Sands
and Desert
Dunes;’ Methuen and Co., Ltd., London, 1954.

W. S. Chepil, “Dynamics of Wind Erosion: 1:’ Soil Science, 60,
305-320, 1945.

W. S. Chepil, “Dyrramics of Wind Erosion:
II ,“ Soil Science,
60,397411,
1945.

W. S. Chepi, “Dynamics of Wind Erosion: ill. The Transport
Capacity of the Wind;’ Soil Science, 60,475480,
1945.

W. S. Chepil, “Dynamics of Wind Erosion: IV. The Transloca-
tion
and
Abrasive Action
of the Wind:’
SoiJ Science, 61,
167-177, 1945.

W. S. Chepil, “Dynamics of Wind Erosion: V:’ Soil Science,
61,257-263,
1946.

J. W. Healy and J. J. Fuquay
“Wind Pickup of Radioactive
Particles from the Ground”
2nd UN Geneva Conference P/391
USA, Pergamon Press, London.

115

.

ATTENDEES

.
Robert E. Allen
Division of Operatio ml Safet y
U.S. Atomic Energy Commission
Washington, DC 20545

L Aoki
REECO
P.O. Box 1440
NTS.-Mercury, NV 89023

Robert E. Baker
Division of Operational Safety
U.S. Atomic Energy Commission
Washington, DC 20545

Capt. William T. Bartlett
USAF Radiological Heafth Laboratory (AFLC)
Wright-Patterson Air Force Base
Dayton, OH 4S433

C. T. Bishop
Mound Laboratory
Monsanto Research Corporation
Miamisburg, OH 45342

Wayne A. Bliss
Environmental Protection Agency
P.O. BOX 15027
Las Vegas, NV 89114

M. R. BOSS
Dow Chemical Co.
Rocky Flats Division
P.O. Box 888
Golden, CO 80401

Andre Bouville
United Nations
New York, NY 10017

Howard Boyd
Lovelace Foundation
5200 Gibson Blvd., SE
Albuquerque, NM 87108

J. F. Bresson
U.S. Atomic Energy Commission
Albuquerque Operations Office
P.O. Box 5400
Albuquerque, NM 87115

L. W. Brewer
Sandia Laboratories
P.O. BOX 5800
Albuquerque, NM 87115

WWiamC. Bright
Dow Chemical Co.
Rocky Flats Division
P.O. Box 888
Golden, CO 80401

C. Austin Burch
U.S. Atomic Energy Commission
Albuquerque Operations Office
P.O. Box 5400
Albuquerque, NM 87115

Percy E. Clemens
Gulf-United Nuclear Fuels Corporation
Research and Engineering Center
Grasslands Road
Elmsford, NY 10523

John P. Corley
Battelle Northwest
P.O. Box 999
Richland, WA 99352

Roger C. Dahbnan
Oak Ridge National Laboratory
Building 3017
Oak Ridge, TN 37830

Jared J. Davis
Office of Effects Evaluation
U.S. Atomic Energy Commission
P.O. Box 14100
Las Vegas, NV 89114

Jack H. Doyle
AECOP-AEC
P.O. Box 5
Oak Ridge, TN 37830

Gordon M. Dunning
Office of Environmental Affairs
U.S. Atomic Energy Commission
Washington, DC 20545

C. E. Elderkin
Battelle Northwest
P.O. Box 999
Richland, WA 99352

W. P. Ellis
U.S. Atomic Energy Commission
Washington, DC 20545

R. J. Engleman
Division of Biology and Medicine
U.S. Atomic Energy Commission
Washington,
DC 20545

B. S. Evans
Colorado Department of Health
42101 lth Avenue
Denver, CO 80220

Gordon C. Facer
Division of Military Application
U.S. Atomic Energy Commission
Washington, DC 20545

William D. Fairman
Argonne
National Laboratory
9700 South Cass Avenue
Argonne, IL 60439

T. R. Folsom
Scripps Institution of Oceanography
University of California, San Diego
P.O. Bos 109
La Jolla, CA 92037

C. W. Francis
Oak Ridge National Laboratory
Building 2001
Oak Ridge, TN 37830

Dave Freas
Battelle Columbus Laboratories
505 King Avenue
Columbus, OH 43201

J. E. Frederick
U.S. Atomic Energy Commission
Washington, DC 20545

Marvin C. Gaske
Advisory Committee on Reactor Safeguards
U.S. Atomic Energy Commission
Washington, DC 20545

Carl C. Gamertsfelder
U.S. Atomic Energy Commission
Washington, DC 20545

Christopher Gatrousis
Lawrence Livermore Laboratory
University of California
P.O. Box 808
L1verrnore, CA 94550

Enc Geiger
Eberlirre Instrument Corp.
P.O. BOX 2108
Santa Fe, NM 87501

A. S. Goldin
Harvard University
School of Public Health
665 Huntington Avenue
Boston, MA 02115

Gerald Hamada
Teledyne Isotopes
4062 Fabian Street
Palo Alton, CA 94303

Stanley E. Hammond
Dow Chemical Co.
Rocky Flats Division
P.O. Box 888
Golden, CO 80401

117

R. W. Loser
Dow Chemical Co.
Rocky Flats Division
P.O. Box 888
Golden, CO 80401

Carol Oen
Ecological Sciences Division
Oak Rid8e National Laboratory
Oak Ridge, TN 37830

Edward P. Hardy, Jr.
U.S. Atomic Energy Commission
New York Operations Office
376 Hudson Street
New York, NY 10014

.

.
Donald Paine
Dept. of Radiology and Radiation Biology
Colorado State University
Fort CoUins, CO 80S21

John H. Harley
Health and Safety Laboratory
U.S. Atomic Energy Commission
376 Hudson Street
New York, NY 10014

Don Majors
Kerr-McGee Corporation
P.O. Box 315
Crescent, OK 73028

WiUiamE. Martin
BatteUe Columbus Laboratories
505 King Avenue
Columbus, OH 43201

Stewart E. Poet
NCAR
P.O. Box 1470
Boulder, CO 80302

A. J. Hazle
Colorado Department of Health
4210 llth
Street
Denver, CO 80220
Jack R. Poison
Mason & Hanger - Silas Mason Co., Inc.
P.o. BOXS61
Burlington, IA S2601

C. H. Mauney
Sandia Corporation
2511 San Mateo, N.E.
Albuquerque, NM 87110

Vernon F. Hedge
Scripps Institution of Oceanography
University of California, San Diego
P.O. Box 109
La Jolla, CA 92037
B. J. McMurray
Atlantic Richfield Hanford Compa
P.O. BOX 250
Richland, WA 993S2

WIIfred L. Polzer
my
U.S. Atomic Energy Commission
Idaho Operations Office
P.O. BOX 2108
Idaho Falls, ID 83401

Edward J. Jascewsky
U.S. Atomic Energy Commission
Chicago Operations Office
9800 South Cass Avenue
Argonne, IL 60439
Rod Melgard
Trapelo/West
2030 Wright Avenue
Richmond, CA 94804

Otto Raabe
Lovelace Foun&tion
5200.Gibson Blvd., SE
Albuquerque, NM 87108
James E. Johnson
E. 1. DuPont SRP
Bldg. 735-A
Aiken, SC 29801
Ctyde Michel
“
Dow ChemicaI Co.
Rocky Flats Division
P.O. Box 888
Golden, CO 80401

WMiam H. Rhoads
EG&G
130 Robin Hill Road
Goleta, CA 93017
H, Kayuhia
REECO
P.O. Box 1440
NTS - Mercury, NV 89023
D. E. Michels
Dow Chemical Co.
Rocky Flats Division
P.O. Box 888
Golden, CO 80401

Chester R. Richmond
Division of Biology and Medicine
U.S. Atomic Energy Commission
Washington, DC 20S45
hlaj. Kurt M. Lanrmers
Kirtland Air Force Base
Albuquerque, NM 87117
D. M. Robertson
BatteUe Northwest
P.O. Box 999
Richland, WA 99352

Ernest J. Lang
Eberline Instrument Corp.
P.O. flex 2108
Santa Fe, NM
87501

Carl R. Miller
Environmental Protection
A&JSCy

Radiation Office
Room 18-B-33, Parklawn BIdg.
RockvUle, MD 20852
Evan Romney
Laboratory of Nuclear Medicine
UCLA
900 Veteran Avenue
Los Angeles, CA 90024

RuaaeULease
REECO
P.O. Box 1440
NTS - Mercury, NV 89023

Raymond L. MiUer
U.S. Atomic Energy Commission
Albuquerque Operations Office
P.O. Box 5400
Albuquerque, NM 87115
Robert Lieberman
Environmental Protection Agency
Southeastern Radiological Heatth Lab.
P.O. BOX 61
Montgomery, AL 36101

John L. RusseU
Environmental Protection Agency
Radiation Office
5600 Fishers Lane
RockviUe, MD 20852

Jofu Mishima
BatteUe-Northwest
622-R Bldg., 200-W. AIea
Richahrd, WA 99352
Carl L. Lindeken
Lawrence Livermore Laboratory
University of California
P.O. Box 808
Livermore, CA 94550

Robert K. Schulz
Dept. of Soil and Plant Nutrition
108 Hilgard HaU
University of California
Berkeley, CA 94720

Robert K. MuUen
EG&G
130 Robin HiURoad
Goleta, CA 93017
.

Craig A. Little
Dept. of Radiology and Radiation Biology
Colorado State Unviersity
Fort CoUins, CO 80521

George Sehmel
BatteUe Northwest
P.O. Box 999
Richland, WA 99352

Randy Newton
Division of Construction
U.S. ATomic Energy Commission
Washington, DC 20545

\

118

.

.

.

J. D. Shaykin
U.S. Atomic Energy Commission
Albuquerque Operations Office
P.O. Box 5400
Albuquerque, NM 87 11S

Warren E. Sheehan
Mound Laboratory
Monsanto Research Corporation
Miamisburg, OH 45342

Col. Buren Shields
FC/DNA, FCWD
Sandia Base
Albuquerque, NM 87115

C. W. SiU
U.S. Atomic Energy Commission
Idaho Oeprations Office
P.O. BOX 2108
Idaho Fds.,
ID 83401

Alvin E. Smith
Atiantic Richfield Hanford Company
P.O. BOX 250
Richland, WA 99352

James Stearns
Dow Chemical Co.
Rocky Flats Divison
P.O. Box 888
Golden, CO 80401

W. L. Stevens
Sandia Laboratories
P.O. BOX 5800
.

Albuquerque, NM 87115

Howard A. Storms
VaUecitos Nuclear Center
General Electric Company
P1easanton, CA 94566

N. A. Tahdtie
Environmental Protection Agency
P.O. BOX 15027
Las Vagas, NV 89114

Charles W. Thomas
BatteUe Northwest
P.o.Box 999
Richland, WA 99352

Michael Thommes
Argonne National Laboratory
9700 South Cass Avenue
Argonne, IL 60439

Michael W. Tiernan
U.S. Atomic Energy Commission
Richland Oeprations Office
P.O. Box 550
Richland, WA 99352

Joseph F. ‘rinney
Lawrence Lwermore Laboratory
University of California
P.O. Box 808
Llvermore, CA 94550

Harold TSO
Eberline Instrument Corp.
P.O. BOX 2108
Santa Fe, NM 87501

W. Twenhofel
United States Geological Survey
Federal Bldg.
Denver, CO 80202

Gerald Ulrikson
Ecological Sciences Division
Oak Ridge National Laboratory
Oak Ridge, TN 37830

AUen M. Valentine
Kerr-MeGee Corporation
Kerr-McGee Building
Oklahoma City, OK 73102

Herbert L. Volchok
U.S. Atomic Energy Commission
New York Operations Office
376 Hudson Street
New York, NY 10014

Stanley Waligora
Lovelace Foundation
5200 Gibson Blvd., SE
Albuquerque, NM 87108.

E.D. Wa8rer
VaUecitos Nuclear Center
General Electric Company
Pleasanton, CA 94566

E. C. Watson
BatteUe Northwest
P.O. Box 999
Richkind, WA 99352

George A. Welford
Health and Safety Laboratoyr
U.S. Atomic Energy Commission
376 Hudson Street
New York, NY 10014

Robert A. Wessman
Trapelo/West
2030 Wright Avenue
Richmond, CA 94804

J. F. Willgirrg
Dow Chemical Co.
Rocky Fiats Division
P.O. Box 888
Golden, CO 80401

Kai M. Worrg
University of California, San Diego
Scripps Institution of Oceanography
P.O. Box 109
La JoUa, CA 92037

Robert E. Yoder
Lawrence Liverrnore Laboratory
University of CaUfornia
P.O. Box 808
Livermore, CA 94550

CM/cb: 500 (212)

119

.

.

.

.

.